sf-3268 forest engineering conference - framsidor till ... · lumber retailer lumber wholesaler...

Uppsala Science park, SE-751 83 UPPSALA, Sweden Ph. +46 18 18 85 00 • Fax. +46 18 86 00

[email protected] • http://www.skogforsk.se

Maria Iwarsson Wide & Berit Baryd (Eds.)

12—15 May 2003, Växjö, Sweden

Posters:DECISION SUPPORT SYSTEM/TOOLSRAW MATERIAL UTILIZATIONORGANIZATION/HRM

Proceedings

Maria Iwarsson Wide & Ingegerd Hallberg (Eds.)

2nd Forest Engineering Conference Posters

3

Practical solution for timber flow control inside a stand

Sakari Suuriniemi Timberjack Oy, Tampere, Finland

Timberjack has been intensively investing in the R&D work of logistic solu-tions in forestry business. Information flow around the forest machines has been increasing all the time. That’s why Timberjack wants to develop software and applications for handling all the information. Fluent flow of information in the logistic chain is the basis of cost effective timber procurement business. By serving tools and systems to handle all the needed information, Timberjack is supporting it’s customers and their activities.

Now Timberjack has also connected forwarders into this logistic chain of tim-ber procurement. It’s possible to use production data of the harvester in the forwarder PC system. On the display of forwarder PC the operator can see where the logs are located on the site. With this information the operator can plan the most effective route of the forwarder and this way prevent nature from excessive driving. Thus the most critical assortments can easily be locali-zed, collected and then transported to the roadside quickly. By sending a pro-duction file from the forwarder, operator can provide forest company with the real time information about volumes of the assortments at the roadside and the coordinates of the storage place(s). With this information transportation of timber from the forest to the mill can be planned most effectively (Picture 1).

Figure 1.


4

Information flow in the harvesting process:

1. Harvester operator starts the site with the new apt-file. Production will be saved in pri-file where every log is stored separately (e.g. diameter, volume and position (WGS 84-coordinates)). Harvester tracks will be recorded with TimberNavi program during harvesting. Pri-file will be sent to for-warder with e-mail automatically according the mail profiles, which has been set for these machines. Situation is updated between machines, which are usually working on the same site. Harvester tracks layer will sent also to the forwarder. (See Picture 2)

2. Forwarder operator can see site borders, storage places and other impor-tant things like key biotopes on the digital map with the TimberNavi pro-gram. After pri-file and harvester tracks layer have been sent from the for-warder, operator can see also position(s) of the assortment(s) and tracks where harvester has been driving. (See Picture 2)

Figure 2.

3. Forwarder operator can activate critical assortment from the assortment list in TimberNavi program and see where that assortment is located on the site. With this information and with recorded harvester tracks, forwar-der operator can optimise the route on the site. (See Picture 3)

4. Forwarder operator loads logs to the load space and then unloads logs to the storage. In this operation operator removes production from active production area (black square) in TimberNavi program and then transpor-ted timber will be displayed in ‘Transported’ field. It’s possible to export prd-file to the forest company about transported volumes and storage place coordinates. (See picture 3)

2nd Posters Forest Engineering Conference

5

Figure 3.

Name: Sakari Suuriniemi (Master of Science, Forestry) Company: Timberjack Address: Box 474, 33101 Tampere, FINLAND Task: Project Manager Department: CTL Heads and Automation Contact: Tel:+358205846922, mobile:+358408475443 E-mail:[email protected]


A didactic experiment: an evaluation of supply chain

management in swedish forest sector context

Bo Dahlin1 and Dag Fjeld2 1South Swedish Forest Research Centre, SLU, Sweden

2Department of silviculture, SLU, Sweden

The literature suitable for teaching forest logistics at a BSc or MSc level is scarce. Therefore, MSc students studying logistics at SLU were given the task to write a review on forestry logistics in general and Supply Chain Management (SCM) in the forest sector in particular (Figure 1). The review maps logistics processes in the forest sector supply chain and examines the possibility for improved effectiveness though implementation of supply chain management principles. The review comprises the backbone of the logistics content in SLU’s post-graduate education in business administration for the forest sector.

Figure 1. The students in the logistics course produced an 84 paged report on Forestry Logistics

This has been repeated for two years. The difference is that the first report was written in Swedish and the second in English. The structure of the two reports also differ somewhat, as does the text, since the authors (the students) are

6


7

completely separate. Differences in lectures, literature and companies visited, also influence the final result.

The first part is an introduction to the basics of logistics terminology and central elements of supply chain management. Examples are given to illustrate how the theory applies to the forest sector.

The next part examines theory and practice around wood supply, processing and distribution. Theories on e.g. storage, transport and information are presented and case studies from Swedish forestry are included (e.g. figure 2).

Figure 2. The flow of material in the forest sector.

Skogen

Virke från skog, köpt virke, importerat virke

The first report (the Swedish speaking one) has a concluding chapter including a holistic perspective of the whole supply chain in order to make a SWOT analysis of potential SCM applications. The second report closes with a summary and conclusions drawn from the studies.

Each chapter was linked to a weekly teaching module that included both theory and case study. Each weekly module was concluded with the draft of the chapter text. In addition, a custom-designed simulation tool (Wood Supply Game) was used to give the students an introduction to potential for the indus-trial dynamics of the forest sector supply chain. The course was concluded with a student seminar where the contents of all chapters were presented.

Although the time for the course was restricted to only four weeks, the results were very good. The experiment also resulted in an improved course evalua-tion from the students. They demonstrated considerable engagement in the

Eget massabruk Eget sågverk

Pappersmassa Virke Flis

Köpt flis

Energi Köpt pappersmassa

Eget pappersbruk Industrikund

Grossist

Konsument Finpapper Tidningspapper Förpackningsvara m.m

Återvinning

Återvinning


8

task, always on their toes in lectures and on visits to find material for the report and working over time to complete the assignment. We consider the problem-based approach as well-suited for this topic because of the greater need for integration of concepts and anticipation of implementation challenges.

References Kulstadvik S., Dahlin B. & Fjeld D. (eds.) 2002. Skoglig logistik – supply chain

management i svensk skogssektor (Forestry Logistics – Supply Chain Management in the Swedish Forest Sector). The Swedish University of Agricultural Sciences, Department of Forest Product and Markets. Report No 4. 84 pp. (in Swedish)

Vestlund, K. (ed) 2003. Forestry logistics handbook. in prep.

Profile Dag Fjeld, lecturer in forest technology, born 63-12-28 BSc in Forestry, University of British Columbia DrSc in Forest operations, Agricultural University of Norway Earlier positions held at Norwegian Forest Research Institute, Danish Forest and Landscape Research Institute. Now at Swedish University of Agricultural Sciences.

Dag Fjeld Department of silviculture SLU SE-901 83 Umeå Sweden +46 90 786 58 56 [email protected]

Bo Dahlin Sydsvensk skogsforskning, SLU Box 49 230 53 Alnarp 040-41 51 81; 070-63 123 61 [email protected]


9

The wood supply game

A Logistics Flight Simulator for the Forest Sector

D. E. Fjeld University lecturer, Swedish University of Agricultural Sciences

Sweden

E. Y. Haartveit Researcher, Norwegian Forest Research Institute

Norway

Keywords: Supply chain structure, industrial dynamics, educational application, forest industry.

Summary Efforts to improve efficiencies in the forest industries have mainly focused on problems within the borders of the company. Thus, there is an unexploited potential for increased efficiency through integrating and coordinating active-ties between companies. This poster presents games that mimic the forest in-dustry and demonstrate to the players how the structure of the system, inclu-ding restricted access to demand information, can result in a mismatch bet-ween supply and demand. The resulting chaos, characterised by rapid transi-tions from situations with considerable backlogs of orders to situations with excessive inventories, makes it difficult for the players to regain control of the system.

Introduction The Beer Game (BG) was developed by Sloan School of Management in the 1960s and is one of the most frequently used student exercises in management and logistics education. The four players are each managing a company in a distribution chain consisting of a brewery for beer, a distributor, a wholesaler and a retailer who delivers beer to consumers.

Transition from Beer to Wood BG is a chain with no branching of the material flow, and where no produc-tion occurs (Figure 1). With the aim of increasing the relevance for logistics in forestry, flows of divergence (such as in log and lumber manufacturing) and convergence (such as mixing several raw materials for paper production) were included in the game structure.


10

The modification of BG into the Wood Supply Game (WSG) involves connec-ting a lumber chain with a paper chain by introducing the forest as a common source of raw materials (Figure 1). A wood supply group diverges the material flow by delivering pulpwood to a paper mill and sawlogs to a sawmill. The paper mill supplies the paper chain, while the sawmill is, in addition to delive-ring lumber products to the lumber chain, also delivering chips to the paper mill. Hence, the mixture of chips and pulpwood for paper production consti-tute a point of convergence, while bucking into sawlogs/pulpwood (wood supply group), production of lumber/chips (sawmill), and production of paper/by-products (paper mill) constitute points of divergence (Figure 1).

Raw materialsConsumer

Retailer Wholesaler Distributor Brewery



Consumer

Consumer

Lumber

Paper

Chips

Wood supply group

PulpwoodSawlogs

By-products

Forest

Paper wholesalerPaper retailer

Lumber retailer Lumber wholesaler Sawmill

Paper mill

Consumer

Consumer

Lumber

Paper

Chips

Wood supply group

PulpwoodSawlogs

By-products

Forest



Paper mill

Beer Game

Wood Supply Game





Consumer

Consumer

Lumber

Paper

Chips

Wood supply group

PulpwoodSawlogs

By-products

Forest



Paper mill

Consumer

Consumer

Lumber

Paper

Chips

Wood supply group

PulpwoodSawlogs

By-products

Forest



Paper mill

Beer Game

Wood Supply Game

Figure 1 The structure of the standard Beer Game and how it is modified in the Wood Supply Game. Each of the games requires one player to operate each company (except “consumer” which is a deck of demand cards).

Methods – Playing the game The game as played in real life is shown in Figure 2. The objective for the players is to minimise the total costs of the chain, and there are only two cost drivers to consider:

A weekly unit cost of keeping inventory.

A weekly unit cost of accumulating a backlog of orders (being out of stock). One unit in the backlog of orders equates to twice the cost of one unit in inventory.

WSG is played for a fixed number of cycles (weeks). Each cycle involves cer-tain activities: Receiving raw materials, order processing, shipping of products, inventory control and placing new orders. Inventories and goods in transit are visible to all players. Order levels, on the other hand, must be kept secret for others than the sender and the receiver. Hence, the retailer is the only player with knowledge of consumer demand. It requires two weeks to transfer orders between players and two additional weeks to ship products back to the custommer. Consumer demand is identical for all competing chains, and is also constant during the game, except for one change occurring early in the game.


11

Figure 2. Students playing the wood supply game. The forest is to the right in the picture, while the consumer markets for paper and lumber products are to the left. Products moves from right to left, while orders moves from left to right.

Results and lessons learned Costs incurred by the different players are distributed unevenly (Figure 3). In WSG, as in BG, costs are higher for positions further away from the markets. The wood supply group has lower costs than the mills. This can be explained by shorter delivery lead time for the wood supply group. Varying costs for the positions (Figure 3), obviously result from differences in the two cost drivers: accumulation of inventory and accumulation of backlog.

0

400

800

1200

1600

Retailer Wholesaler Mills Wood supplygroup

Position

Cos

ts (U

SD) Beer Game

Wood Supply Game

0

5

10

15

20

25

Stan

dard

dev

iatio

n of

or

der

rate

s

Retailer Wholesaler Mills Wood supplygroup

Position

Beer Game

Wood Supply Game

Figure 3. How costs (left) and the standard deviation of order rates (right) are for the different positions in WSG (N=12), compared to the corresponding positions in BG (N = 11).

Changes in inventory Studying one particular game, inventories for all positions are rather stable the first couple of weeks, and start shrinking for all positions after 5-8 weeks (Figure 4). Backlogs (negative inventory) are most notable for the sawmill, followed by the wholesaler and the wood supply group. Inventories for each position result from placed orders and order pattern upstream in the supply chain.


12

-200

-150Week

Inve

ntor

y

-100

-50

0

50

100

150

1

Lumber RetailerLumber WholesalerSawmillWood Supply Group

Lumber RetailerLumber Wholesaler

0

50

100

Week

Size

of o

rder

SawmillWood Supply Group

Figure 4. Development of inventory and the corresponding development of the size of order for one parti-cular WSG.

The placed orders have the highest peaks for the wood supply group and the sawmill (Figure 4). During the post game debriefing session, most players comment on the huge variation in the orders they placed, and blame this on a volatile market. However, during a WSG market demand never exceeds 10, and only changes once. Evidently, the system itself was responsible for most of the dynamics observed.

Based on this, the observed variation of order rates is amazing, but well known in situations where several actors in a sequence make their decisions based on incoming orders from their immediate customers. Systems characterised by restricted access to information (on consumer demand), and time lags between neighbouring actors often experience that demand amplifies upstream in the supply chain. One relevant measure of demand amplification is the standard deviation of the order rate. It is obvious from Figure 3 and Figure 4 that the standard deviation of order rates is increasing upstream in the supply chain.

Although demand is identical, the development of the two chains in a WSG often differs. This causes problems in the points of divergence and conver-gence. For example, the sawmill might face huge demand for lumber and no demand for chips, thus increasing chips inventory. For the point of conver-gence, the problems occur when one of the vendors fails to deliver one of the raw materials (pulpwood and chips) that are both required for paper pro-duction. The increased complexity and dependency leads to reduced perfor-mance for WSG compared BG (Figure 3).

The key role of inventories is to decouple material flows, and thereby reduce their inter-dependency. Recent trends in the forest industry include attention to inventory reductions in order to reduce costs. These savings may, however, become difficult to realise unless operations are considered jointly and infor-mation on supply and demand are released to customers and suppliers. In a supply chain with small inventories and high degrees of dependency, there is a risk of increasing the variation of demand. Effects of demand amplification are stronger towards companies upstream in supply chains; the wood based industries and forest operations (Figure 3).


13

Integrating transport costs In the most resent version of the game three chains (WSGs) are integrated through a common source of raw materials, where the nodes in the network illustrate supply points in the forest (Figure 5). Transport costs depend on the distance from the supply point to the chain at issue. In situations where supply exceeds demand, wood can be procured from desired locations with moderate transport costs. However, when demand exceeds supply additional transport costs apply in wood procurement. Pilot experiments using this larger version of WSG indicate that the chains with the highest variances of order rates are also penalised with increased unit costs of transportation. This compares well with wood procurement in real life, as procurement costs tend to increase in situations with limited supply.

Figure 5. In the latest version of the wood supply games, competing supply chains are integrated by being supplied form the same forest.

Conclusion WSGs have been used in the last three years in topics of logistics in the fores-try education at the Swedish University of Agricultural Sciences (SLU) and at the Agricultural University of Norway (NLH). The complexity of WSG has continuously increased, and the latest version occupies 30 students for 3 hours. The games convincingly demonstrate to the players, whether students, mana-gers or researchers, how actors in supply chains are constantly affected by each other’s decisions, and that lack of communication can cause serious difficulties for the actors concerned.

Profile Dag Fjeld, lecturer in forest technology, born 63-12-28 BSc in Forestry, University of British Columbia DrSc in Forest operations, Agricultural University of Norway Earlier positions held at Norwegian Forest Research Institute, Danish Forest and Landscape Research Institute. Now at Swedish University of Agricultural Sciences.


14

Erlend Ystrøm Haartveit. I grew up in Aremark municipality, just on the right side of the border to Sweden. I have an MSc in forestry from the Agri-cultural University of Norway (AUN). Presently, I am employed as a researcher at Skogforsk (Norway), and am working on problems concerning logistics in forestry, mainly in the solid wood industries. I am also pursuing a PhD within the same topics. My logistics education is from Scandinavian universities, and from the University of British Columbia, Canada where I had a wonderful stay in 2000/2001. Finally, I also work as a lecturer (logistics in forestry) for the (AUN).

Dag E. Fjeld Erlend Ystrøm Haartveit Swedish University of Agricultural Sciences Umeå, Sweden

Norwegian Forest Research Institute, Ås, Norway

+46 907865856 +47 64949095 [email protected] [email protected]


15

Efficient and flexible wood supply strategies in the Lithuanian forest

sector – a comparison of lead times, delivery precision and demand

amplification for sawlog delivery and processing

Mindaugas Puodziunas Agricultural University of Lithuania, Faculty of Forestry

Department of Forest Management, Lithuania

Dag Fjeld Swedish University of Agricultural Sciences, Faculty of Forestry

Department of Silviculture, Division of Forest Technology, Sweden

Summary In this study evaluation of supply chain strategies are done and estimation of lead times and delivery precision are made for roundwood deliveries from two different forest enterprises in Lithuania. The two enterprises represent case studies of efficient and flexible supply chain strategies.

The demand amplifications are examined for total 6 state forest enterprise sawmills, which are classified according to high, medium and low economic potential. The evaluation if supply chain performance for Lithuanian sawmills is linked to economic result is also made.

The results showed that shortest lead times were found for the high value veneer assortments (22 days). It was followed by sawlogs (39 days), pulpwood (49 days) and palletwood (58 days). The study shows the shortest lead times for the high value assortments and species, which are more in focus for the enterprise with the flexible supply chain strategy. Four of 10 (40 %) customers or the efficient enterprise purchased within 20 % of their agreed volumes. The corresponding number for the flexible enterprise was 2 of 10 (20 %).

The analysis of data demonstrated that inventory cover times for high and medium classes of sawmills for roundwood and sawnwood were the same and reached 10 and 12 days, respectively. In low economical potential sawmills the corresponding figures were 14 and 29 days. For the mills producing high pro-portions of high quality assortments the coefficient of variation for final sales varied between 5 and 35%. The bullwhip factor was greatest for sawmills with shortest inventory cover times.

Keywords: Supply chain strategy, timber market, industrial dynamics.


16

Introduction The timber markets in Lithuania have been greatly affected by the political transition during the past decade. Now, after many years of the forest sector being almost invisible to other European countries Lithuania has re-entered the European wood markets. The situation in the state sawmilling sector, however, is still difficult. The new market conditions require radical changes for survival in this sector.

During the Soviet-period, timber sales were mainly been driven by silvicultural requirements, pushing timber on the market. This is called the biological forestry where the main production function concerned a maximisation of volume. Now in the new situation, wood supply is becoming more and more market-oriented where current demands pull timber into the market (Heinemann, 2000).

In other sectors, varying production approaches are classified as supply chain (SC) strategies. Fisher (1997) characterizes two such strategies for manufactu-ring Industries: innovative and efficient. Lehtonen (1999) adapts this dichoto-my to the paper sector and characterizes two relevant variants: flexible and efficient. In Lehtonen’s dichotomy high utilisation of production facilities, even-flow of materials, standard product range and poor customer service levels are common features of the so-called efficient strategy. This is a strategy which still prevails in much of the Nordic and Lithuania forest sector, resulting in slow response to market fluctuations. In turn, the flexible supply chain strategy is characterised by an uneven-flow of materials and a wider product range, and where high customer service levels play a major role in companies’ performance (Lehtonen 1999).

The potential consequences of demand amplification in industrial supply chains, termed industrial dynamics, were first analysed by Forrester (1958). Forrester (1958) demonstrated how small variations in final customer demand may be amplified upstream in the supply chain, initiated by slow order hand-ling, lack of downstream sales information and immediate corrective actions for inventory discrepancies. Finnish researches Hameri and Nikkola (1999) were first who studied these effects in supply chains for papermills. According to Hameri (1996) the demand for products is transmitted along the series of inventories using stock control ordering and estimates.

Aim To evaluate efficient and flexible SC for Lithuanian sawmills is linked to economic result.

Materials and Methods A case study for lead times and delivery precision has been carried out in two state forest enterprises. The data collection was made for year 2001.

In manufacturing sectors lead times are defined as the time from the order is given to when goods are delivered to the customer. It includes the sum of re-planning time and time spent for harvesting, storage and transportation. In forestry this variable shows how fast the enterprise is


17

able to adjust the flow of roundwood to the customer. In the experiment 30 stands for each enterprise were studied including 15 stands with typically high demand (e.g. high value birch) and 15 stands with typically low demand (e.g. spruce pulpwood).

Delivery precision is defined as the ratio between delivered volume and planned volume for the order period of time. This measure shows how accurately the supplier/customer trade follows the agreed sales plan. The data was analysed using selling agreements and monthly sales reports.

The statistical analysis of lead times was done in the SAS statistical package. The data was analysed with the General Linear Models (GLM) procedure. Differences were considered significant at p<0.05.

In the next study supply chain performance for 6 state forest enterprise sawmills was examined. Within the list of sawmills classified according to high, medium and low economic potential. Monthly product flows were collected for each sawmill. Three indicators were estimated per assort-ment. The first is inventory cover time, (ICT). The second is the coeffi-cient of variation (CV= / ) for received roundwood (R), processed sawnwood (P), and sawnwood sales (S). The third is the degree of corre-lation for monthly changes in upstream and downstream flow.

In particular the ratio of variation upstream (CVup) to flow variations down stream (CVdown) is indicative of problems of demand amplification or supply variability. The ratio between these is called bullwhip effect ( ) and is measured as either = CVup/Cvdown

Results Lead time The average leads were on average 45 days for efficient and 42 days for flexible enterprise. In a complete statistical model (Table 1) studied factors included: assortment (assort), species (spp), enterprise (ent) and combinations of assort-ment–species (assort*spp) and assortment–species–enterprise (assort*spp*ent). The enterprise was the only factor that was not correlated with a statistically significant difference.

Table 1. ANOVA table for lead times: factors included are assortments (assort), species (spp) and enterprises (ent). Source Type I SS Mean Square F Value Pr>F assort 2384491,647 596122,91 48,75 <. 0001 spp 544034,242 77719,177 6,36 <. 0001 ent 21907,948 21907,948 1,79 0,1812 assort*spp 382292,617 38229,262 3,13 0,0006 assort*spp*ent 764818,771 50987,918 4,17 <. 0001 The results showed that shortest lead times for assortments were found for the veneer. The average lead time for this assortment was 22 days. It was followed by sawlogs (39 days), pulpwood (49 days) and palletwood (58 days). The comparison of lead times among assortments for the two studied enterprises is showed in Figure 1.


18

50

19

38 4050 49 52

66

0

20

40

60

80

y sda

v eneer s aw logs pulpw ood palletw ood

Figure 1. The length of the lead times for different assortments in the enterprises. First column represents efficient and second flexible enterprises.

Delivery precision The analysis of delivered volumes in relation to sales agreements per quarter showed average values for the efficient enterprise fluctuating between –29,1% (2nd quarter, 2001) to –12,3% (3rd quarter, 2001). For the whole period of 2001, actual sales fell short of agreed sales by 12 673 m3 (-17,8%). The situation in the flexible enterprise had different features. The best periods, in terms of fulfilling sales agreements were the 1st and 4th when precision was –1,7% and + 6,7%, respectively. These quarters however were interspersed poorer precision during the 2nd and 3rd quarters when the actual sales were only 50 % of agreed sales. This situation led to the negative result at the end of year when actual sales fell 21,9% short of agreed sales.

Demand amplifications A comparison of the proportion of processed volume to palletwood and other higher value products (unedged and edged boards) showed that the percent of total production volume consisting of high value products clearly varied bet-ween the different mills as well as inventory cover times also varied consider-ably between following factors (Table 2) such as: enterprise (high, medium and low), wood type (roundwood and sawnwood) and product group (high and low value).

Table 2. Inventory cover times for the different assortments of the six studied sawmills.

sawmill number

high value palletwood volume

(%) ICTr

(days) ICTs

(days) volume

(%) ICTr

(days) ICTs

(days) 1 80 n/a 13.5 20 n/a 5.1 2 74 9.52 25.6 26 10.5 8.1 3 36 8.86 14.7 64 11.1 11.6 4 62 7.03 9.8 38 9.41 11.4 5 48 10.7 52.7 52 13.8 13.7 6 18 39.9 52.3 82 6.26 13.4

The variation of inventory cover times among three classes of sawmills is shown in Figure 2.


19

0HIGH MEDIUM LOW

5

10

15

20

25

30

35

days

Figure 2. The inventory cover times in different classes of sawmills. First column represents roundwood and second represents sawnwood The ratio between the coefficients of variation for roundwood deliveries and sawn wood sales ( ) however was generally higher for the high quality assortments than palletwood (Table 3).

Table 3. Coefficients of variation (CV) and bullwhip factors ( ) for the different assortments of the six studied sawmills. sawmill number

high value palletwood CVs p/s r/s CVs p/s r/s

1 0.152 0.89 n/a 0.315 0.86 n/a 2 0.24 1.81 2.24 0.277 1.36 1.27 3 0.155 1.26 2.05 0.218 1.08 1.91 4 0.168 1.25 1.74 0.262 0.62 0.82 5 0.446 0.75 0.90 0.218 0.93 1.67 6 0.676 1.24 1.09 0.575 1.04 0.52

Conclusions 1. There is no significant difference in average lead times. Lead times for

higher quality timber were twice shorter for flexible enterprise compared with efficient one.

2. Delivery precision both in efficient and flexible enterprises was low and during studied period fell short by 17,8% and 21,9% and in terms of biggest customers by 14.8 % and 33.8 %, respectively.

3. Inventory cover times for roundwood and sawnwood was equal in high and medium economic potential sawmills – 10 and 12 days, respectively, whereas in low economic potential sawmills corresponding figures was 14 and 29 days. Nevertheless the ICV was higher for sawnwood compared with roundwood in all studied sawmills.

4. The smallest bullwhip effect was detected between sales and processing while between sales and received roundwood it was the greatest. This effect was higher for high quality assortments compared with low value ones.


20

References Fisher M. L. 1997. What is the right supply chain for your product? Harward Business

Rewiev 1997, 2:105-116.

Forrester J. W. 1958. Industrial Dynamics. MIT Press, Cambridge, MA, 1961: 497 p.

Hameri A.P. 1996. The method to avoid demand amplification in bulk paper distribution. Paper and timber journal, Vol. 78(3): 102-106.

Hameri A.P. and Nikkola J. 1999. Demand amplifications in supply chains of mills close to and far from main markets. Paper and timber journal. 1999, Vol81/No.6: 434-438.

Heinemann H. R. 2000. Business process re-engineering – a framework for designing logistics systems for wood procurement. In: Sjostrom, K. (ed.), 1st World Symphosium on Logistics in Forest Sector. Helsinki 2000: 269-287.

Lehtonen H. 1999. Supply chain development in process industry. Acta Polytechnica Scandinavica IM No. 4.

Profile My name is M. Puodziunas and I am holding MSc degree in Forestry. Currently I am doing my PhD studies in Lithuanian University of Agriculture and Swedish University of Agriculture Sciences. The study field is forest logistics. This topic covers the investigation of wood flow from forest to a customer. The activities under this scope include harvesting techniques, transportation, storage and processing of wood. Different supply chain strategies are estimated and their adaptability to Lithuanian forest sector are analyzed.

I am also involved in the certification of Lithuanian forest enterprises and I have participated in several forest management assessments. Mindaugas Puodziunas Agricultural University of Lithuania, Faculty of Forestry, Student 13, Akademija, Kauno raj., Lithuania +370 698 46425 [email protected] Dag Fjeld Swedish University of Agricultural Sciences, Faculty of Forestry Department of Silviculture, Division of Forest Technology, 90183 Umea, Sweden +46 90 786 58 56 [email protected]


21

Usage of computer simulation for planning of wood transport under mountainous conditions of Ukraine

Oleg Styranivsky Assoc. Prof., Ukrainian State University of Forestry and Wood Technology

Ukraine

Summary For opening up of the forest area it is necessary to judge the efficiency of the existing wood transportation system, to choice environmentally benign type of logging method and of wood skidding mean. So many factors and alternatives are considered and evaluated in this process that it creates conditions where the necessity is to use modern computerized methods.

The aim of work is an elaboration of computer based analysis for planning of route at perspective forest road network, as part of Special Decision Support System for appropriation of forest massif to logging.

The proposed algoritm for placement of forest road route on digital model is based on methods of forest road designing at mountainous conditions on maps of non-digital format. For the placement of perspective forest road route it is analysed as informational layers, that passes traditional information on topographic terrain peculiarities, so layers with digital information on future perspective usage of forest road for non-forestry types of managment.

Keywords: Transportation planning, route of forest road, digital terrain model, GIS environment.

Introduction For opening up of the forest area it is necessary to evaluate existing forest road network, to choice technology method and means suitable for harvesting and transportation of wood. Transportation planning is quite complicated task. It is necessary to take into considerations abundant number of indices, factors and criteria, which are to be met for reaching the aim. It takes necessary to use GIS-system with additional software modules what makes possible to carry out different multivariate, situational and aim-oriented calculations

The aim of work is an elaboration of computer based analysis for planning of route at perspective forest road network, as part of Special Decision Support System for appropriation of forest massif to logging.


22

Materials and Method A lot of articles are describing planning models and criteria for evaluation of forest road network (Kobayashi, H. 1984, Yoshimura, T. 1997, Tucek, J. Pacola, E. 1999). As a vulnerable moment of the most of them it is a lack of complex approach to determination of final aim. Because incomes in future could be obtained from exploitation of forest road network only under its complex usage for aims forestry and non-forestry types of activities (farming, recreation, tourism…) (Yoshimura, T. Styranivsky, O. Bybljuk, N. 2002). In view of it for planning of perspective forest road network together with traditional databases (digital terrain model, existing road network, descriptions of forest and wood skidding means) it has to be taken into account the information on perspective sites for tourizm, recreation, hunting, farming, grasing… (Potocnik, I. 2002).

It is proposed to study this approach for wood transportation planning on the example of forest area at Rozhanka forestry unit of Slavsk state forestry enterprise. By evaluation results of existing wood transportation system at test forest area it was substantiated the necessity of extension of existing forest road network. After detailed site studies, taking into account some historical events, it was formed Arc View GIS database on distribution of cultural and historical sites, sightseeing sites, former hunting bungalows, perspective routes for summer and winter tourism (Styranivsky, O. 2002).

Description of Algoritm for Route of Forest Road Modelling The proposed algoritm for placement of forest road route on digital terrain model is based on methods of forest road designing at mountainous conditions on maps of non-digital format (Treskinsky, S. A. 1974). At same time it is necessary to take into account information about slope direction and distri-bution of water flows, relief transformations, objects of forestry, tourism and recreation managment. But at GIS software environment it is not so easy to evaluate the form of landscape - slope direction (uphill or downhill?), place-ment of significant natural lines (ridge or meadow?). Thus, as a first step for placement of forest road route on digital model we are proposing to show its basical direction, taking into account as the slope of terrain so the distribution of significant objects of forestry, tourism and recreation managment.

The next step for every raster cell of digital model it is determined index K of its suitability for placement of forest road route which is equal to sum of slope index Y and remoteness index B. The method of identification for suitability indices K and B is illustrated in Figure 1. All informational data necessary for further analysis are presented in form of separated layers of cell map model.


27

21,6

16,2

10,8

5,45

10

15

20

25

30

500

400

300

200

100

00

100

200

300

400

500

600

0 1 2 3 4 5

remoteness index

dist

ance

to b

asic

al

dire

ctio

n, m

000 1 2 3 4 5

slope index

slop

e, d

egre

es

Figure 1. Relations of suitability indices to real data

Identification of route direction is starting from initial cell of forest road basical direction and it is carried out for every next the most suitable cell of map model in such sequence: out of all cells of map model it is selected the number of cells neighbouring to the cell, which was chosen at previous step (their altitude indices are not lower and they situated not farther to remote cell of basical direction); out of selected number are chosen cells suitable to futher analysis by suitability index ( 5 ) and by slope degree (Figure 2); the most suitable cell for prolongation of route direction of forest road is determined by conditions: =max or an excess of altitude index =min (for y cells); paragraphs 1–3 are repeated up to the moment of impossibility of their carrying out.

23

90

Figure 2. Selection by slope degree of cells suitable for analysis: - cell, selecten at previous step; – cell, suitable for analysis; – cell, which could be selected for analysis in case of unsuitable of x cells.

In Figure 3 it is presented the example of placing of forest road route with usage of above mentioned algoritm. There is presented the line of forest road route placed manually. As we see, deviations between manual and computer variants are insignificant, what it is stressing on insignificant deviations by proposed algoritm.

$T;

ÊÚ

suitability index5 - 5.55.5 - 66 - 6.56.5 - 77 - 7.57.5 - 88 - 8.58.5 - 99 - 9.59.5 - 10

project route (by manually)project route (by computer)

$T tops; former hunting bungalows

streams and riversremnants of old historical roadpathspublic road

ÊÚ forestry officebasical direction of forest road route

0.8 0 0.8 1.6 Kilometers

N

EW

S

Figure 3. Project route of forest road on digital terrain model

90 < 180 180 < 270


24

To the forest road network of test area it was added two project routes and after it was done analysis of its effectiveness. As it was predicted, the percentage of forest area, that is not covered by transportation network has been reduced from 9 % to 0,3 %. Presented data prove the effectiveness of carried out extension of forest road network on test area.

Discussion Proposed in the article the algoritm of forest road route simulation on digital terrain modell is presented as a tool software GIS environment for help of engineers which are dealing with design of forest road to carry out swift and effective planning of transportation system, what at final end has influence on possibility to use environmentally benign technology of wood harvesting. Algoritm uses valuable feature of GIS software product to create a wide range on initial data in form of separate informational layers. For the placement of perspective forest road route it is analysed as informational layers, that has traditional information on topographic terrain peculiarities, so layers with digital information on future perspective usage of projected forest road for non-forestry types of managment.

References Kobayashi, H. 1984. Planning system for road-route locations in mountainous forests.

Journal of the Japanese Forestry Society 66(8): 313-319.

Yoshimura, T. 1997. Development of an expert system planning a forest road based the risk assessment. Ph.D. Thesis. Kyoto University. 82 p.

Tucek, J. Pacola, E. 1999. Algorithms for skidding distance modeling on a raster digital terrain model. Journal of Forest Engineering 10(1): 67-79.

Yoshimura, T. Styranivsky, O. Bybljuk, N. 2002. Planning ecotourism in the Carpathians. Naukovyj Visnyk 12.1: 198-203.

Potocnik, I. 2002. Categorization of forest roads on the basis of their multiple use and social needs. Naukovyj Visnyk 12.1: 66-73.

Styranivsky, O. 2002. The substantiation of wood skidding methods in the mountainous conditions of the Carpathians. Proceedings of the international conference "Logistics of wood technical production in the Carpathian Mountains". Zvolen: 244-249.

Treskinsky, S. A. 1974. Gornye dorohi. Transport. 368 p.

Profile Dr. Oleg Styranivsky is an Associate Professor of Ukrainian University of Forestry and Wood Technology, Ph.D. degree since 1987. Research area: plan-ning of wood transportation under mountainous conditions, environmenttally sensitive technologies for wood harvesting and transportation. Published more than 50 research papers and completed several research projects.

Oleg StyranivskyGeneral Chyprynka str.103, UkrSUFWT 79057 Lviv Ukraine +380 322 390 669 (work) +380 679 503 849 (mobile) [email protected]


25

Planning of long-distance seedling transportation

J. Rantala Researcher, Finnish Forest Research Institute

Suonenjoki, Finland

Summary In this paper, planning methods for long-distance seedling transportation were studied. To determine the optimal transportation costs in different strategies for seedling production, linear programming (LP) and mixed integer program-ming (MIP) were applied. To manage spatial information, a geographical information system (GIS) was used. As a result, the relative improvement in cost-effectiveness caused by centralized transportation system compared to decentralized transportation system varied from 13.0 % to 36.5 %, depending on the production strategy used. Applicability of LP to planning of long-distance seedling transportation was compared to that of MIP. In the LP model transportation costs were determined per seedling whereas in the MIP model they were based on vehicle loads. With relatively large number of seed-lings included in optimization, standard LP seems adequate method for strate-gic planning of seedling transportation in Finland. The MIP model seems to be appropriate tool for smaller, operative level, optimization problems.

Keywords:Transportation problem, optimization, nursery business, linear programming, mixed integer programming, Finland.

Introduction In Finland, the number of nurseries has recently been decreasing year by year, and it seems probable that in the near future this trend will continue. Simul-taneously the number of forest owners’ associations, which are the most im-portant middlemen in seedling business between nursery companies and forest owners, has decreased too. For these reasons, together with hardened compe-tition between nursery companies and general pressure on improving the cost-efficiency of silvicultural operations, the importance of transportation planning in seedling business has increased. In addition to the arguments mentioned above, it can be assumed that planting through the whole growth period will become more common in the future, which again emphasizes the importance of careful planning of seedling transportation.

Methods In this paper, transportation planning options and methods for long-distance seedling transportation were evaluated in different production strategies of a nursery company. The production strategy consisted of the number of nursery units, production allocation among nursery units and transportation allocation


26

among time periods. The options compared were centralized and decentralized transportation planning systems; in centralized system all orders from nurseries owned by the nursery company were gathered together and optimized as a whole, whereas in the decentralized system each nursery unit carried out its’ own seedling delivery (Rantala et al. 2003). In practice, decentralized system leads to a situation where internal transportation is needed between nursery units of the company. From the methodological standpoint, the applicability of linear programming (LP) to management of seedling transportation was com-pared to that of mixed integer programming (MIP). In the LP model, tran-sportation costs were determined per seedling, whereas in the MIP model they were based on vehicle loads (Figure 1, Rantala 2003).

Average number of transported vehicle loads per transportation route

Tran

spor

tatio

n un

it co

st

MIP (theoretical worst solution for ntransportation routes *)MIP (single transportation route *)

LP

* all routes include 1 customer

Figure 1. Principle unit cost functions for the LP and MIP models.

Results and Conclusions The current development towards seedling transportation managed by nursery companies seems to have marked advantages in the cost-effectiveness of trans-portation; the relative improvement in cost-effectiveness caused by centralized transportation strategy compared to decentralized transportation strategy varied from 13.0 % to 36.5 %, depending on the production strategy used. Naturally, a decrease in the number of nursery units raised the total transporta-tion cost due to an increase in the total transportation distance. The increase in costs was much smaller when centralized system was applied to transportation planning instead of traditional decentralized system (Figure 2). In Fig.2, it should be noted that a decrease in the number of nursery units means only putting an end to production, so that the original 5 units will continue as sales and storage sites.


27

0

20

40

60

80

100

120

140

160

Cos

t ind

ex

Number of nursery units

External transp. Internal transp.

5 13

(i)

(i)(i)(ii)

(ii)

(ii)

Figure 2. Effect of the number of nursery units on costs of long-distance transportation in centralized (i) and in decentralized (ii) transportation system. Effect of production allocation among nursery units on transportation costs was studied in production strategy of 3 nursery units. The increase in specia-lization among nursery units slightly emphasized the advantages of centralized transportation system (Figure 3).

0

20

40

60

80

100

120

140

160

Degree of product specialization

Cos

t ind

ex

External transp. Internal transp.

Medium(3-5)

High(2-3)

(i) (i)(i)

(ii) (ii)

(ii)

None(5)

Figure 3. Effect of product specialization (number of seedling types produced) on long-distance transportation costs in centralized (i) and in decentralized (ii) transportation system.

When the number of transported seedlings within a certain period decreased, for instance, due to planting through the whole growth period (PTGP), the computational accuracy of the LP model was clearly lower than that of the MIP model. Despite that, differences in allocation of orders between the models studied were small; in a case of 5 nursery units, which is the current production strategy of the company studied, only few orders were allocated to another nursery unit when the MIP model was applied instead of the LP model. In production strategies of three or less nursery units no difference between the models occurred. The MIP model took remarkably more time to solve with up-to-date PC than the LP model. Thus, in the actual business


28

situation of Finnish nursery companies, standard LP seems to be an adequate tool for management of seedling transportation. Correspondingly, the MIP model seems appropriate tool for smaller, operative level, optimization problems such as seedling transportation of singular nursery unit or singular transportation period.

In PTGP transportation was divided either in 3 or 5 periods. Contrary to the LP model, the MIP model proved practical when the effect of PTGP on transportation costs was evaluated. From the standpoint of the cost-efficiency of the seedling transportation, PTGP increased optimal minimum costs markedly. The transportation costs increased 10 – 20 % depending on the numbers of transportation periods and nursery units. As the number of seedlings transported during a certain period decreesed, transportation unit costs increased exponentially (Figure 4). The increase in transportation unit costs seems to be very slight until the number of transported seedlings decree-ses to less than 12,000,000; from then on, the increase in costs is strongly accelerated.

80

90

100

110

120

130

140

150

160

170

0 4 8 12 16 20 24

Number of seedlings (millions) transported during a certain transportation period

Tran

porta

tion

unit

cost

inde

x

Figure 4. Effect of the number of seedlings transported within a single period on transportation unit costs. The optimal proportion of utilization of the vehicles with smaller transporta-tion capacity increased while the total number of seedlings transported on a certain transportation route during a certain period decreased (Table 1).


29

Table 1. Effect of the number of nurseries and transportation periods on optimal allocation of transportation for a truck with a trailer and a pick-up truck.

Number of nursery units

Number of transportation periods

Truck with a trailer Pick-up truck

1 1 96.02 % 3.98 %

1 3 86.95 % 13.05 %

3 1 95.70 % 4.30 %

3 3 84.06 % 15.94 %

3 5 77.01 % 22.99 %

5 1 94.13 % 5.87 %

5 3 78.36 % 21.64 %

These results will be useful for nursery companies and forest owners’ associations when they evaluate the cost effects of production allocation, product specialization and systems of transportation management. This presentation is a summary of two articles mentioned in the references.

References Rantala, J., Kiljunen, N. and Harstela, P. 2003. Effect of seedling production and long-

distance transportation planning strategies on transportation costs of a nursery company. Will be published in International Journal of Forest Engineering 14(2).

Rantala, J. 2003. Linear programming (LP) and mixed integer programming (MIP) in management of seedling transportation. Submitted Manuscript.

Profile Name: Juho Rantala

Education: M.Sc.(For.) -01, Forest technology and wood industry (University of Joensuu)

Title: Research Scientist

Organisation: Finnish Forest Research Institute, Suonenjoki Research Station

Expertise: Regeneration technology, nursery logistics

Current research project: Technology, organizing and logistics of plantation forestry

Juho Rantala Finnish Forest Research Institute, Suonenjoki Research Station FIN-77600 Suonenjoki, Finland +358 17 513 8317 +358 50 391 4855 (work) [email protected]


30

Fuel consumption of forest machines and timber trucks

Kaarlo Rieppo, MSc (Eng, For) & Jouni Väkevä, MSc (For) Organization: Metsäteho Oy, P.O. Box 194,

FIN-00131 Helsinki, Finland

Summary Information of fuel consumption is needed to make cost and environmental calculations and also to minimize the fuel consumption and emissions caused by forest machines and timber trucks. The main goals of the research are to find out the average consumption levels, the main factors affecting these consumption levels and their respective influence. Another goal is to develop the measuring methods.

The project is going on from autumn 2002 to spring 2003. The co-operators are the Jyväskylä Polytechnic Institute, EC-Tools ltd, the Ministry of Transport and Communications and FinnRA.

The level of the fuel consumption of harvesters and forwarders is measured by mechanical Piusi-flow meters. Measurements are taken every time the tank of a machine is filled. The data is collected from 24 harvesters and forwarders in-cluding different sizes of Valmet, Ponsse and Timberjack machines. The research period of each machine is 1–2 months. Also the volume of processsed timber and data about the working circumstances are collected.

More detailed information about fuel consumption is gathered from one machine by a Motec-data collecting system, which allows for measurements to be taken at a maximum rate of one per second. Detailed data makes it possible to carry out the analysis in a new way and to find the main factors affecting to the fuel consumption.

The fuel consumption of timber trucks is measured by monitoring vehicle and driver performance. Data is collected automatically via a ”black box” and transmitted wireless to database and reported via www. Main purpose is to show fuel consumption of a vehicle and some interesting performance indica-tors like usage of brakes, rpm, average speed etc. The location of the vehicle is monitored by GPS and the X-, Y- and Z-coordinates are combined to the other information. Summary lists of results can be made from a specific time period. All indicators are also shown detailed in a trip level. The monitoring solution is called as EC-ROAD. The product is originally built for the use of truck operating companies, but the methodology makes it possible to carry out the research work in a modern way.

A pilot study is being carried out simultaneously on how to utilize vehicle motion simulator (VeMoSim). The simulator is based on dynamics (laws of physics) and the technical characteristics of the vehicle (engine, power train and gear ratios, drive resistances etc.), vertical and horizontal geometry of the road and the driving technique. In this study the most interesting outputs are


31

the fuel consumption and the pollutant emissions (NOx, CO, HC, PM and CO2). Simulation makes it possible to analyze how e.g. the driving technique, type of vehicle or the road characteristics affects the fuel amount and emissions.

Profile – Jouni Väkevä Research topics timber transportation

forest roads

cost calculations

profitability of enterprises

History

Masters of science in forestry 1991, University of Helsinki

Researcher in the Finnish Forest Research Institute 1991 - 1997

Researcher in Metsäteho Oy 1997 -

Contact information

Kaarlo Rieppo, MSc (Eng, For) Organization: Metsäteho Oy, P.O. Box 194, FIN-00131 Helsinki, Finland [email protected]

Jouni Väkevä, MSc (For) Organization: Metsäteho Oy, P.O. Box 194, FIN-00131 Helsinki, Finland [email protected]


32

A SDSS to compare a wide range of harvesting scenarios on the basis of

wood procurement cost, wildlife habitat and land use indicators

Osvaldo. Valeria E.1 Luc G. LeBel2

Kim Lowell3 1PhD Candidate, 2Associate Professor,3Professor

Université Laval, Canada

Summary A spatial decision support system (SDSS) was designed to help forestry planners take into consideration aboriginal land uses in eastern Canada. This SDSS permits the comparison between different harvest scenarios based on four criteria: wood procurement, accessibility, wildlife habitat and land use. Each criterion is evaluated through a set of indicators calculated by mathema-tical models. The SDSS operates as an ArcGis8.x extension and allows one to measure spatial parameters used to calculated economic and land use indicators.

An economic model is used to evaluate the operational cost of a given harvest scenario using productivity and cost functions for four sub models: planning, road, harvesting and transportation. The economic model takes into considera-tion harvest patterns, harvest systems (tree length or cut-to-length), equipment displacement, road classes, block sizes and seasonal effects.

The land use model generates information regarding wildlife habitat index (moose, marten and rabbit), area accessible from forest roads, forest cover changes, and the impacts on sensitive zones. These indicators were identified by the aboriginal community members as being important to assess the impacts of industrial forest management on their traditional way of life.

Finally, the indicators for each scenario are reported in a decision matrix. The decision matrix permits forestry planners and aboriginal community members to compare scenarios for economic and non-economic products.

Keywords: GIS, forest operations, decision support.


33

Introduction Several communities in Canada depend largely on forest companies to create local employment and generate economic activities. Native people may not have benefited as much from the activities of forest products companies as others, and forest harvesting can disturb their traditional activities. While con-frontations have marked the issue of forest management in the past, forest products companies are now working with local groups to try to accommodate non-fibre-related uses of the forest. Recently, alternatives to current harvesting procedures have been sought in an effort to reduce the impacts of forest operations on non-fiber-related activities.

A native-owned and operated harvesting corporation (Mishtuk) and the Waswanipi Cree Model Forest (WCMF) in the Abitibi region have been using variable patch size forest management (VPSFM) as their preferred intervention procedure. The key to VPSFM is harvesting the forest in patches of the size and shape that best correspond to ecological and societal criteria for the region. A private company is in a trial phase of VPSFM management near Waswanipi. Although their mosaic pattern is more regular than the one used on the WCMF and their cut blocks are slightly larger, this company continues to contend with similar problems: higher operational costs and time and labor consuming planning efforts.

Currently, there is no integrated methodology for conducting forest planning based on VPSFM that allows for an a priori evaluation of the effect(s) of a given management plan on economic and non-fibre-related values. In the absence of the necessary tools to evaluate such a management approach, the process of public consultation for forest management may become unnecessa-rily frustrating and adversarial.

In response to this situation, a project was carried out to develop a geographic information system (GIS)-based decision support tool. The project was con-ducted in cooperation with one forestry company, Mishtuk, and representa-tives of the WCMF.

Objectives – The primary objective of this project was to develop a spatially based

forest management decision support tool that is designed specifically to provide information on the financial and non-economic consequences of a given forest management plan on variable patch size forest management (VPSFM).

– The secondary objectives were:To increase understanding of the cost associated with managing the boreal forest in patches of varying sizes with respect to values of non-fibre-related forest values.

– To facilitate the capacity of forest companies to develop innovative forest management strategies to provide financial profit while respecting non-fiber-related values.

– To provide a mechanism by which all groups that use forest resources can feel that they are part of the forest management process.


34

Literature review Decision support for forest management and planning activities has undergone two revolutions since computers began to become more commonplace in the late 1960s. The first was a quantitative revolution; among the domains that initially benefited from this increased computing power were forest growth and yield modelling (e.g., Sullivan and Clutter 1972, Ek 1974), and forest planning optimisation through operations research (e.g., Stuart 1981, Bullard et al. 1985). The question to which an answer was sought from quantitative tools of this era was generally “How much?” (“How much wood will be on the forest in 10 years if I employ a particular silvicultural treatment?” “How much will it cost to extract 1000 m3 from the forest?”)

The second revolution was not quantitative, but instead was spatial: “Where?” became the fundamental question rather than “How much?”. (“Where can I harvest economically?” “Where do I need to undertake reforestation?”) The increasing power and decreasing cost of GIS software fed this spatial revolu-tion. As a result, studies seeking to optimise forest planning activities while considering spatial constraints began to appear on such diverse subjects as road planning (Nelson and Brodie 1990) and harvest scheduling (O’Hara et al. 1989). Non-fibre-related forest uses such as wildlife habitat (Baker et al. 1995) and watershed management (Schloss and Rubin 1992) have also been studied in a spatial context.

All of the studies cited in the previous paragraph made use of the capacities of a GIS to examine spatially related topics in isolation of each other. However, because of its analytical capabilities, GISs have also been used to integrate rese-arch derived from various research domains into a single management tool (e.g., Brown et al. 1992). The benefits of using a GIS for such a purpose are not confined to analytical capabilities, however. Visualizing and communica-ting the effects of various forest management strategies is also one of the strengths of a GIS (Orland 1994). Another important benefit is the capacity to pose “what if?” questions and to evaluate the results of a variety of different management scenarios.

The VPSFM approach presents challenges and opportunities in the province of Quebec as well as in other regions of the boreal forest. In this type of forest, harvesting systems have not been designed to be excessively mobile and flexible. As stated by Rummer et al. (1997), ecosystem management will foster a need for harvesting systems that can be economically viable for various silvi-cultural methods. Therefore, given the actual equipment and contractor base, perhaps the greatest challenge for procurement foresters will be to minimise ever-increasing harvesting and road construction costs. Several models dealing with cut-block size and spatial distributions have acknowledged higher capital spending and maintenance costs for roads (Nadeau, 2002) and cost of equip-ment and labour movements (Zundel 1992, O’Hara 1989). Cost reductions dictate that close attention is given to movements between cutting blocks and road network layout.

The development of a viable wood procurement strategy requires that harves-ting cost be minimised. But at the same time, decision-makers must also attempt to maximise the positive externalities of their strategies. Models have been designed to assist decision-makers in their choice of a harvesting solution that will best fit their needs. For example, Wang (1994) provided a decision


35

support system (DSS) that, based on case studies from northeast China, can be used to monitor multi-stage inventories in forest harvesting while taking into account environmental, economic, and safety considerations. Other tools such as REMSOFT’s Stanley (Remsoft 1996) can help planners design their tactical harvest blocking/scheduling. However no application was identified that could rapidly evaluate a wide range of harvesting scenarios based on wood procure-ment cost jointly with the impacts on forest usage. The underlying objective of the present study is to use VPSFM as an opportunity to improve forest planning by taking into consideration a wide range of impacts on the environ-ment.

Methods The first step in this project was to assemble information from “knowledge-able” sources for input into the proposed GIS-based decision support tool. Most data was collected at the Waswanipi Cree model Forest (WCMF). The territory is typical of boreal forest found across Quebec. Company records on operation costs were obtained. Meeting with Cree representatives allowed the identification of important issues related to traditional land use by the com-munity. Wildlife data and indicators were obtained through meetings with Crees and biologists knowledgeable of the region. The objective at this stage was to learn what critical information is required to properly compare diverse harvesting scenarios.

Once the data had been obtained, the challenge was to develop a process that considered all relevant input and variables to compute indicators for a set of key criteria. It is on the basis of these criteria that scenarios can later be compa-red. The general structure of the DSS is schematically represented in Figure 1.

Forest management scenarios are generalized on the basis of a set of spatial parameters and operational, social, and biophysical variables. These, in turn, are used as inputs defining each scenario. The inputs are processed to provide quantitative information for a set of four criteria. A brief description is made of these criteria and indicators. Details about the calculations and formulas used to compute these indicators can be found in the software user guide (Valeria et al. 2003).

Procurement costs

Wildlifehabitats

Accessibility

Land use

Procurement costs

Wildlifehabitats

Accessibility

Land use

InputsInputs

Process Process

Variables Scenarios Scenarios

Selection Selection

Results

Criteria

Figure 1. General structure of the SDSS.


36

Procurement costs Wood procurement costs are computed through an economic model that incorporates productivity and cost functions. The model takes into considera-tion four subsystems : planning (timber cruising, delimiting blocks), road network (construction, maintenance), harvesting activities (harvest, moving) and transportation (hauling, loading). The user must define the area where operations will be conducted. Based on the spatial information managed with the GIS, most costs are automatically computed through various algorithms. Harvesting and skidding productivity can be modelled using FERIC’s func-tions available on Interface (Favreau & Cormier, 2000). Alternatively, user defined parameters allow for the estimation of a production function based on local knowledge.

An innovative harvest blocks clustering approach was developed to simplify spatial measurements of variables affecting procurement costs. Adjacency rules, spatial indices, and operational factors are used to write scripts that per-mit the identification of operational clusters (Figure 2). Wood procurement costs are computed for three hierarchical spatial scales: cutting blocks, cluster and area of interest.

Wildlife habitat Several indicators are used to evaluate wildlife habitat. The first set of indicators presents the distribution of forest stands by age class and cover type. This provides an indication of the forest landscape after harvest. The second set includes quality habitat indices for moose, marten, and rabbit. Finally, the number of continuous mature forest cover blocks is computed.

Figure 2. Clusters of harvesting blocks with their access roads.

Accessibility Two indicators are evaluated to report on accessibility. The percentage of the area accessible by road is measured using a buffer zone of a user-specified width. The proportion of sensitive areas made accessible because of harvest is estimated by tallying the number of sensitive areas intersected by the previously defined buffer zone. Sensitive areas are defined by local users and can take several forms: camp, burial ground, river landing, etc.


37

Land use Land use is reflected through seven indicators: 1) total harvested area, 2) total harvested volume, 3) number of cutting blocks, 4) number of clusters, 5) total road distance, 6) number of sensitive areas overlapped by at least one cutting block, and 7) total harvested area and area recently harvested.

Results All the scripts required to evaluate the above mentioned criteria were coded in Visual Basic. The result is a software called the Wood Procurement Planning Tool (WPPT) that works in conjunction with a GIS. To install the application, the following are required: a ArcGIS/ArcView 8.1 or higher licence; a Spatial Analyst licence; Microsoft Windows 2000 or XP with the latest service pack installed; Microsoft Excel.

Once the user has provided the required spatial and numerical information, the software computes values for the selected indicators of all four criteria. Preliminary results are displayed through a series of forms (Figure 3).

Figure 3. Forms generated by WPPT to provide quantitative information for each criteria.

Using WPPT, forest planners can efficiently interact with other stakeholders in the area to prepared wood procurement scenarios that meet their respective objectives. The best-suited scenarios can then be evaluated and compared using a decision matrix that outlays critical decision-making variables


38

(Figure 4). The matrix is intended to summarize the results and should help decision makers establish the tradeoffs of selecting one option over the others.

Figure 4. Decision matrix used to compare competing scenarios.

Conclusion WPPT is a decision support tool that is effective in comparing forest manage-ment scenarios on the basis of procurement costs, wildlife habitat, land and accessibility. This spatial tool is currently tested in a “real life” situation in the boreal forest of northern Quebec. Results indicate that it is useful to foresters who have to must prepare forest management plans in situation where local population has to be consulted.

Because the software runs as an extension to ArcGis8.x it is relatively simple to integrate it with other spatial planning aids. In its current form, WPPT does not attempt to find an optimum harvesting strategy. There is no immediate plan to develop such functionality in the near term. Rather, plans are underway to provide a more complete set of indicators in regards to hydrology and soil impacts after harvest. Also, improvements in the financial model could provide more flexibility to consider in more details the long-term costs of each scenario.

Acknowledgments The authors are grateful for the financial support provided by the Natural Sciences and Engineering Research Council of Canada (NSERC), the Canadian Forest Service, Abitibi-Consolidated inc., and the Waswanipi Cree Model Forest. Access to FERIC’s production functions was also greatly appreciated.

References Baker, B., Cade, B., Mangus, W., McMillen, J., 1995. Spatial analysis of sandhill crane

nesting habitat. Journal of Wildlife Management 59:752-758.

Brown, J., White, C., Holcroft Weestra, A., Wedgwood, R., 1992. Use of the SPANS GIS in the development of an ecosystem management model for the Central Canadian Rockies Ecosystem. Proceedings: The Canadian Conference on GIS, March, Ottawa, Ontario, p. 258-265.


39

Bullard, H., Sherali, H., Klemperer, W., 1985. Estimating optimal thinning and rotation for mixed-species timber stands using a random search algorithm. Forest Science 31:303-315.

Ek, A., 1974. Nonlinear models for stand table projection in northern hardwood stands. Canadian Journal of Forest Research 4: 23-27.

Favreau, J. & D. Cormier, 2000. Interface récolte/harvesting 2000. Institut Canadien de Recherches en Génie forestier (FERIC).

Nadeau, F.- R. 2002. Analyse de l’impact de la dispersion des aires de coupe sur les coûts d’approvisionnement en matière ligneuse à la Forêt Montmorency. Mémoire de maîtrise. Université Laval. 133 p.

Nelson, J., Brodie, J., 1990. Comparison of a random search algorithm and mixed integer programming for solving area-based forest plans. Canadian Journal of Forest Research 20: 934-942.

O’Hara, A., Faaland, B., Bare, B., 1989. Spatially constrained harvest scheduling. Canadian Journal of Forest Research 19:715-724.

Orland, B., 1994. Visualization techniques for incorporation in forest planning geographic information systems. Landscape and Urban Planning 30:83-97.

REMSOFT. 1996. Design and development of a tactical harvest blocking/scheduling tool- Stanley. Final Rep. Can. For. Serv., Pac. For. Cent.

Rummer, B., Baumgras, J., McNeel, J. 1997. Forest operations for ecosystem management. In: Barbour, R. James; Skog, Kenneth E., eds. Role of wood production in ecosystem management: Proceedings of the Sustainable Forestry Working Group at the IUFRO all division 5 conference; 1997 July 7-12; Pullman, WA. Madison, WI: U.S. Department of Agriculture, Forest Service, Forest Products Laboratory: 46-52

Schloss, J., Rubin, F., 1992. A “Bottom-Up” approach to GIS watershed analysis. Proceedings: GIS/LIS ’92. November, San Jose, California, p. 672-679.

Stuart, W., 1981. Harvesting analysis technique: a computer simulation system for timber harvesting. Forest Products Journal 31: 45-53.

Sullivan, A., Clutter, J., 1972. A simultaneous growth and yield model for loblolly pine. Forest Science 18: 76-86.

Valeria, O., L. LeBel & K. Lowell. 2003. Wood Procurement Planning Tool version 1.1 User’s Guide. Université Laval, Québec, Québec. 45 p.

Wang Lihai. 1994. Optimal operation planning for integrated forest harvesting and transport operations from the forest to the mill. Journal of Forest Engineering. 6(1):15-22.

Zundel, P.E. 1992. Planification des choix et de l’affectation de systèmes d’exploitation forestière à moyen terme. Doctoral dissertation. Université Laval, Québec, Canada.


40

Profile Luc LeBel is associate professor of forest operations at Laval University in Québec City. He received a master’s degree in Engineering Administration, a Master’s of Science, and a PhD from Virginia Polytechnic and State University. His main research interests are contractor performance evaluation and wood procurement systems analysis. He has worked continuously with logging contractors in the southeastern United States and Canada since 1991. He is also active in the area of wood transportation and forest road network analysis.

Dr. Luc LeBel Faculté de Foresterie Université Laval Québec (Québec Canada G1K 7P4 Phone-number 1.418.656.2131 extension 8835 (work) [email protected]


41

Harvest planning J. Karlsson1, M. Rönnqvist2, J. Bergström3

PhD. Student, Linköping University (LiU), Professor, LiU and The Forest Research Institute,

M.Sc. Forestry, The Forest Research Institute Sweden

Summary The problem we consider is harvesting planning on annual and operative (4-6 weeks) level. The main decisions deal with which harvest areas to harvest, when to harvest and to allocate a harvest team. The harvesting of areas is plan-ned in order to meet industrial demand. The total cost includes harvesting, transportation and storage. Areas are of varying size and the composition of assortments in each area is differrent. Each harvest team has different skills, a different home base and different production capacity. Another aspect is the road network. There is a cost related to road opening (restoring, snow remo-val). The weather conditions vary a lot during the year. Some areas and roads are not accessible during break frost periods and heavy rain periods, due to loose ground. This is an important aspect in annual harvest planning. Opera-tive harvest planning include to decide the harvest sequences or schedules for harvest crews. One important aspect is cost due to the quality reducetion of logs stored at harvest areas. We develop two different mixed integer program-ming (MIP) models for these problems. The annual planning problem can not be solved directly using a commercial MIP solver. We have developed an optimization based heuristic solution method, which produce high quality solutions within practical time limit. With a limited number of schedules, the operative problem can be solved by a commercial MIP solver. The models and heuristics are tested in case studies at Holmen Skog, a major Swedish forest company.

Keywords: Operations Research, harvesting, heuristics, integer programming, modelling.

Introduction The problem we consider is harvesting planning from the perspective of Swedish forestry companies. In comparison with competitors on the interna-tional market, Sweden has large costs due to the infrastructure. This fact force the provision of forest products to be both efficient and customer oriented. Large forest firms own a considerable part of the forest. Larger forest compa-nies have an organisation where the operations are divided into smaller regions, which may be composed of one or several districts. At each district harvesting plans are made for both long term and short term. On a tactical level an overall plan for one year are made. Through a central planning, the annual industrial demand for paper-, pulp- and sawmills are distributed on a monthly level for each district. On an operative level a more detailed plan are made for a couple


42

of weeks. On a short time horizon the industrial demand are determined on a weekly level. The plans are continuously updated.

The harvesting planning include decisions on which areas to harvest and when to harvest so that the industries receive required amount of assortments. It also includes decisions to buy additional areas from smaller forest owners, contrac-ting of harvest teams and planning opening of roads that are necessary for harvesting and transportation. As there are many restrictions and a huge amount of information that is used, it is often difficult to get a clear overview of the overall planning situation. There are most often a limited number of persons that make the plans, and they spend a large amount of time to come up with qualitative plans. Optimization makes it possible to model the complex connections and many restrictions that must be considered. In the Chilean Forestry Industries there are different planning tools based on optimization in use. The system OPTICORT (Epstein et al, 1999a,b) deals with short-term harvest planning. The PLANEX system (Epstein et al 1999) determines the optimal location of towers, roads and skidder operations.

We have developed a mathematical model for the harvesting planning problem at a district, on both tactical and operative levels. We have done this in a hier-archical structure where tactical and operative harvesting plans are made by two different models. The tactical model gives data to the operative. To achieve compatible decision making on different levels, different approaches have been used. Hierarchical methods have been developed to integrate tactical and strategic planning, (Weintraub and Cholaky, 1991). McNaughton, Rönnqvist and Ryan (2000) present an integrated model, which incorporates both strategic long-term goals and detailed area-sensitive tactical planning.

Problem Description The harvesting plans are made start from a pool of areas possible to harvest. When tactical plans are made the pool of areas corresponds to at least 1,5 years harvesting. For each month harvesting areas should be chosen so that the industries demand can be fulfilled. Possibilities of varying the production level are described in Brunberg (2001). This involves e.g. to select areas with larger trees, areas with particular assortment mix, allow overtime, change or add equipment for crews. When short-term plans are made an overall plan should be available, describing suitable areas for the specific planning period. Each harvest area is unique with particular properties. Areas are of varying size and the composition of assortments is different. Each area is either fully harvested or thinned. Once it is harvested it produces a given amount of assortments. The produced amount per day is depending on middle size of the trees.

Harvesting planning include decision of which harvest team to use for each area. Each team has different skills. That means the teams have equipment suitable for thinning or final felling. There is a limit on their capacity con-cerning middle size on the trees. The teams have different production capa-cities. Another important thing to consider is the road network. The harvesting planning includes decision about opening of roads connecting the areas to public roads to make harvesting and transportation possible. In winter there are costs for opening roads due to snow removal. During break frost period restoring can be necessary to make some roads accessible. To control when and which roads that must be open an overall transporttation plan are made.


43

The planning problem also includes control of storage levels in forest, at termi-nals and industries. A significant cost is due to quality deterioration of products stored outdoors at harvest areas or at terminals. At the industries the storage capacity are limited.

Annual Planning In the annual planning all decisions are taken on a monthly level. Harvesting areas are chosen for each month. Their relative order are not considered, which means that costs connected to moving equipment can not be taken into account. The cost due to value decrease of products stored outdoors is con-sidered and estimated per month. The weather and road conditions are varying a lot during one year and affect the annual plan. The accessibility to areas and roads during the different time periods are of vital importance. Certain roads can not be used during some time periods and others should be avoided. Some areas can not be harvested during break frost periods due to loose ground.

Operative Planning The short-term planning includes the same aspect as the annual planning, with a few differences. The total planning period of five weeks contains decision on a weekly level. It is not necessary to consider the accessibility of areas and roads, as the possible areas are assumed to be suitable for the particular plan-ning period. The actual sequence for each harvesting team is defined. The costs associated with moving equipment are taken into account. In a more detailed planning on an operative level it is of vital importance to take in consideration the storage level and the age of the harvested timber. The quality deterioration cost increase for each week of storing. In the short term planning both storage and age are estimated on a weekly level.

Results Both models are linear, mixed integer large scale problems. In the annual planning problem with a pool of areas consisting of 1,5 year harvesting, the number of possible areas is about 400. In our case study, the number of permanent working teams at the district are 5, there are 5 assortments, 5 industries and 12 time periods. That gives a problem with 25 000 binary variables, 220 000 continuous variables and 60 000 constraints. It is not possi-ble to solve the problem directly to optimality within reasonable time limit using some commercial program. Harvesting problems often includes discrete decisions about specific stands or roads, which corresponds to integer variab-les. The presence of integer variables and the size of real-world problems often requires heuristic methods to achieve practical solution times. Nelson and Brodie (1990) give a comparison of a heuristic with an optimization method applied on a 3-decade tactical harvesting problem. Different heuristic methods have successfully been used to different harvesting planning problem, see Weintraub et al (1995) and Yoshimoto et al (1994). We have developed a heuristic that gives high quality integer solutions in about six hours. The operative model involves about 40 possible harvesting areas, 6 working teams, 5 assortments, 5 industries, 5 time periods and 5 age classes of storage at a typically district. We generate a limited number of harvesting sequences. That give a proble with 32 800 continuous variables, 6500 constraints and each


44

possible sequence corresponds to a binary variable. With the number of harvesting sequences limited to 12 000 this model are directly solvable to optimality within distinct time limit.

References Brunberg, T. 2001. Flexibel drivning på Östgården hos Norrskog. The Forest

Research Institute of Sweden, Working paper 472-2001, (in Swedish).

Epstein, R., Morales, R., Seron, J., Weintraub A. 1999. Use of OR Systems in the Chilean Forestry Industry. Interfaces 29(1), 7-29.

Epstein, R., Weintraub, A., Chevalier, P., Gabarro, J. 1999. A system for short term harvesting. European Journal of Operational Research 119, 427-439.

McNaughton A., Rönnqvist, R., Ryan, D. 2000. A model which integrates strategic and tactical aspects of forest harvesting. Proceedings of System modelling and optimization Methods, Theory and Applications, 19th IFIP TC7 Conference on System Modelling and Optimization, July 12-16, 1999, Cambridge, UK.

Nelson, J., Brodie, J. D. 1990. Comparison of random search algorithm and mixed integer programming for solving area-based forest plans. Canadian Journal of Forestry Research 20, 934-942.

Weintraub, A., Cholaky, A. 1991. A hierachical approach to Forest Planning. Forest Science 37:2, 439-460.

Weintraub, A., Jones, G, Meacham, M., Magendzo, A., Magendzo, A., Malchuk, D. 1995. Heuristic procedures for solving mixed-integer harvest scheduling-transportation planning models. Canadian Journal of Forest Research 25(19), 1618-1626.

Yoshimoto, A., Brodie, J. D., Session, J. 1994. A new heuristic to solve spatially constrained long-term harvest scheduling problems. Forest Science 40, 365-396.

Profile – Jenny Karlsson Scince 1999 Jenny Karlsson is a PhD student at Linköping University, Department of Mathematics, Division of Optimization. She is graduated from Linköping University, Master of Science in Mechanical Engineering, 1995. During 1995-1999 she was working at Saab Aerospace, as a Structural Engineer. In 2002 she finished her Licentiate Thesis in optimization. This was dealing with optimization models and solution methods for some forestry applications. The main part dealt with Harvest Planning. This application gives large scale integer models. Her current efforts are focused on other related applications within forestry. Jenny enjoys almost all sports, and especially running, golf and long-distance ice-skating.

Jenny Karlsson Department of Mathematics, Optimization Linköping University 581 83 LINKÖPING email: [email protected]


45

Distributed wood procurement planning within a multi firm

environment Daniel Beaudoin

Ph.D candidate, Université Laval, Quebec, Canada

Luc Lebel Professor, Faculty of Forestry and Geomatics,

Université Laval, Quebec, Canada

Jean-Marc Frayret Associate Director of Research, Research Consortium in e-Business in the forest

products industry (FOR@C), Université Laval, Quebec, Canada

Summary In this poster, the harvest and distribution problem where multiple firms share the same procurement areas is tackled. Each firm correspond to a financially independent entity pursuing local objectives. In such inherently decentralized context, conventional optimisation techniques are inadequate and result into inequities between firms. An automated negotiation approach is proposed to solve this problem while taking local objectives in consideration.

Keywords: Harvest and distribution planning, multi firm environment; distributed planning, automated negotiation.

Introduction Optimization techniques and heuristic methods have been commonly used for solving harvest and distribution problems. Some recent examples include Gunnarsson and al. (2001) and Karlsson and al. (2002a and 2002b). In a centralized context, where planning is carried out in a centralized location by a single agent, such approaches are adequate. However, they are no longer appropriate when planning occurs within a multi firm environment. In this regard, Wightman and Jordan (1991) have demonstrated that solutions obtain with optimization techniques are inequitable when no common goal is shared by the firms. Thus, the overall goal of this project is to develop a decentralized approach to wood procurement planning in a multi firm environment, which will take local objectives into consideration.


46

nn

min

Context In Quebec, more than 85% of the productive forested area is on public land. This area is divided into 161 distinct zones called “common areas”. The government awards contracts to forest companies allowing them to remove, on a yearly basis, a predefined volume of one or more species on a specified common area. A mill may have contracts on more than one common area, and several contracts may be awarded to different companies on the same common area, even for the same species. Sharing a procurement area means that in a single block, harvested volumes are sorted according to their characteristics and their ability to be manufactured into certain product types (for example: softwood lumber, hardwood lumber, pulp, veneer and posts) in order to be delivered to the appropriate mills.

The government requires that long, medium and short-term plans be submitted for each common area. Although every firm is responsible for its own procurement planning, plans must be jointly submitted and requires that all activities be integrated within a common area.

Furthermore, even if most companies manage their own operations, part of their needs for raw materials must be fulfilled through the purchasing of wood from other companies’ operations. These relationships characterized by dependency must be managed by coordinating procurement activities.

Problem The problem tackled is a harvest and distribution problem in a decentralized environment. The main issue arise from the presence of several financially independent entities, all trying to satisfy their particular demands at the lowest possible cost. Since

jj

jj CC

11.min.

where: Cj = procurement cost of mill j, conventional optimization methods cannot find an optimal solution that will satisfy all entities.

In such setting, no single planner has the authority to impose his or her own decisions on others. Planners must come to an agreement on which type and what volume of wood should be harvested each week and which beneficiary should carry out the operations. They must also decide in which blocks to harvest and what type of equipment to use. Also, what volumes should be hauled and to which mills, must be decided as well as the location where the volumes will be stored and for how long.

The CHALLENGE is thus to come up with a plan that will be satisfactory for all.


47

Proposed approach It is propose to solve the procurement planning problem with a two step approach:

Local optimization.

Coordination of partial plans.

Each mill performs a local optimization to identify their most favorable plan. Conflicts will arise for the same resources or timing of the activities. Local optimal plans are the starting positions or offers for the automated negotiation process. Because mills do not share a common goal, coordination of partial plans is accomplished through negotiation. Figure 1 shows the decentralized nature of planning and the required interactions. Since planners are considered rational, no one would accept to move away from its optimal plan to accommodate others unless financially compensated for doing it. To resolve conflicts on resource allocations, monetary compensations are offered. Planners rely on decision models for decision making. These models are based on utility theory to make sure negotiation converge to an agreement.

Figure1. Planning in a decentralized mode. Conclusion The proposed decentralized planning approach is aimed at generating equitable plans from the firms’ point of view. To our knowledge such approach as never been attempted in wood procurement planning. Expected results include redu-ced overhead costs resulting from the time spent on planning and negotiation, and lower local and global procurement costs compared to the actual planning process.


48

References Gunnarsson, H.; Lundgren, J.T.; Rönnqvist, M. 2001. Supply chain modeling of forest

fuel. Department of Mathematics, Linkoping University, Sweden. Report LiTH-MAT-R-2001-08.

Karlsson, J., Rönnqvist, M., Bergström, J. 2002a. Annual harvest planning, Department of Mathematics, Linköping Institute of Technology, Sweden. Report LiTH-MAT-R-2002-15.

Karlsson, J.; Rönnqvist, M.; Bergström, J. 2002b. Short-term harvest planning including scheduling of crews, Department of Mathematics, Linköping Institute of Technology, Sweden. Report LiTH-MAT-R-2002-14.

Wightman, R.A.; Jordan, G.A. 1990. Harvest distribution planning in New Brunswick. Can. For. Ind. 110(2):19-22.

Profile Daniel Beaudoin After graduating in Forestry from Laval University (B.Sc.A in Forest Opera-tions, 1996), I worked in central British-Columbia for a five-year period. During that period I occupied several positions providing me with experience in a wide range of domains. I obtained a M.B.A in 2002 and am actually a PhD candidate in Laval University (Quebec, Canada). I’m an active member of the Association of British Columbia Professional Foresters (ABCPF) and «Ordre des ingénieurs forestiers du Québec» (OIFQ), the equivalent organisation for Quebec’s professional foresters. Luc LeBel Luc LeBel is associate professor of forest operations at Laval University in Québec City. He received a master’s degree in Engineering Administration, a Master’s of Science, and a PhD from Virginia Polytechnic and State University. His main research interests are contractor performance evaluation and wood procurement systems analysis. He has worked continuously with logging contractors in the southeastern United States and Canada since 1991. He is also active in the area of wood transportation and forest road network analysis.

Daniel Beaudoin Cité universitaire, Pavillon Abitibi-Price Ste-Foy, Québec, G1K 7P4 Phone: (418)656-2131, ext.8329 Email: [email protected]

Luc LebelCité universitaire, Pavillon Abitibi-Price Ste-Foy, Québec, G1K 7P4 Phone: (418)656-2131, ext.8835 Email: [email protected]

Jean-Marc Frayret Cité universitaire, Pavillon Adrien-Pouliot Ste-Foy, Québec, G1K 7P4 Phone: (418)656-2131, ext. Email: [email protected]


49

Life cycle assessment and environmental product declaration

of forestry related products and processes – a way to meet environmental objectives

Athanassiadis Dimitris, PhD in Forestry,

Swedish University of Agricultural Sciences Department of Silviculture, Division of Forest Technology, Umeå Sweden

Summary The demand for quantified and verified information about the environmental performance of products and services constantly increases. A number of environmental tools that can help companies to understand and measure the environmental impacts associated with their products, processes, and activities has emerged e.g. Design for the Environment (DfE), Life Cycle Assessment (LCA) and many more. LCA is a systematic set of procedures for compiling and examining the inputs and outputs of materials and energy and the associa-ted environmental impacts directly attributable to the function of a product or service system throughout its life cycle. The aim of the study was to describe the environmental performance of a forwarder crane in a life cycle perspective taking into account the following stages of the crane's life time; (1) raw material acquisition and intermediate processing (2) fabrication of individual compo-nents and assembly of the crane, (3) associated transports and (4) use. It was found that a forwarder crane will consume 1050 MWh of energy during its lifetime while at the same time 360 tonnes of CO2 equivalents (CO2, CH4, etc) will be released into to atmosphere. 97% of that amount of energy and 98% of the CO2 equivalents were due to the use phase. The results were used to pre-pare an Environmental Product Declaration (EPD) for the forwarder crane. With the help of EPD:s, products that perform the same function can be com-pared to each other with respect to energy and resource consumption and associated emissions to air, water and soil, tied to system performance and final product produced.

Keywords: Life Cycle Assessment, Environmental Product Declaration, Forwarder crane, Emissions to air.


50

Introduction Cranab AB commissioned the Swedish University of Agricultural Sciences (SLU), to carry out an LCA for one of its forwarder cranes. The aim of the commissioners was to acquire a certified environmental product declaration (EPD). The company was the first manufacturer concentrating on forestry cranes to be approved in accordance with the international ISO 14001 environ-mental standard and is also EMAS registerred. The EPD system is an attempt to apply ISO TR 14025 (a normative technical report for provisional use in the field of Type III environmental declarations) in practice. The system is based on LCA, according to ISO 14040 – 14043 standards (International Standards Organization, 1999). The company aimed to use the results from the LCA in product development; that is identify opportunities to improve the environmental performance of the product at various points in its life cycle.

The aim of the study was to describe the environmental performance of a forwarder crane based on a life cycle assessment by taking into account the following stages of the life cycle of the crane; (1) raw material acquisition and intermediate processing (referred to in the tables as raw material) (2) fabrica-tion of individual components and assembly of the crane (refereed to in the tables as fabrication), (3) associated transports (raw material to component manufacturers, components to assembly factory, complete crane to final customer) and (4) use.

Materials and methods The CRF 8 forwarder crane was examined. The technical life expectancy of the crane was set to 1 200 000 crane cycles and the production capacity of a for-warder with the CRF 8 throughout its lifetime was set to 680 000 m3 over bark (vob). The reference flow (functional unit) of the study is 1000 m3 vob at the roadside. All energy and materials consumed as well as emissions to the envi-ronment were normalized to the functional unit. The functional unit is crucial as it defines a product’s performance and enables comparisons between EPD:s.

The main components of the forwarder crane are the slewing motor, the pillar, the main boom, the outer boom, the grapple, the hydraulic cylinders, the rotator, and the hydraulic hoses (Figure 1). These are either manufactured at Cranab or at suppliers and transported to Cranab (mainly by truck). The material composi-tion of the crane was decided by examining each component. The contribution of the materials to the total mass of the crane was the following (in kg); Steel/iron: 1960, bronze: 1.2, brass: 7, nitrile rubber: 2, Polyethylene (High Density): 3, Ethylene Propylene Diene: 21, polypropylene 5.


51

Figure 1. The CRF 8 forwarder crane.

Environmental data (amount of input materials, energy consumption, solid waste generation and air/water pollutant release) were collected for all life cycle stages of the crane.

The crane contains at least 15% of recycled steel. For that reason, environ-mental benefit from recycling of the crane was left out of the boundaries of the study. Several, but not all, of the manufacturing processes at the suppliers were taken into account. Use phase modelling included fuel consumption and hyd-raulic oil consumption (3 l/1000 m3 vob) by the forwarder, fuel consumption for the transport of the forwarder between logging sites and parts replacement requirements (mainly hydraulic hoses) of the crane (25 kg steel and 20 kg Ethy-lene Propylene Diene/1000 m3 vob). Forwarder fuel consumption strongly de-pends on vehicle size, engine power, travel distance, load size, log and bunch size, grapple volume, terrain conditions and operator skill. In the case of the 14 ton forwarder with the CRF 8 fuel consumption was estimated to 9.5 l per hour. 70% of that was judged to be due to the crane.

The emission data were classified into five environmental impact categories such as global warming, eutrophication etc. The impact was quantified with the aid of a category indicator. For the global warming environmental impact the category indicator is the global warming potential (GWP), expressed as CO2 eq., of each greenhouse gas (e.g. CH4 equals to 21 CO2 eq., N2O equals to 310 CO2 eq. etc.). The environmental impacts and associated category indi-cators that were used are those recommended by the Swedish Environmental Management Council (2000).

Results Carbon dioxide (CO2), nitrogen oxides (NOx), sulphur oxides (SOx), hydrocar-bons (HC) and particle emissions are some of the main emissions that cause a great impact to the environment and are mostly determined by the amount of energy consumption. The amount of these emissions from the different life cycle stages of the crane are shown in Table 1.


52

Table 1. Amount (g/1000 m3 vob) of selected emissions for the life cycle stages of the forwarder crane.

Life cycle stages CO2 NOx SOx HC Particles Raw material 9 500 11 11 0.2 7.5 Fabrication 1150 1.5 2 10 1 Transport 138 1.2 0.1 0.2 0.2 Use 530 000 4 600 200 550 320

Most environmental impacts occur at the use stage owing to the emissions from fuel combustion. The material acquisition and intermediate processing stage follow the use phase in terms of amount of energy consumed and contri-bution to the global warming impact category. Similar observations were made in a previous study on a forwarder (Athanassiadis 2000). None of the in-house and at suppliers manufacturing stage processes involved a lot of emissions except the painting process where some solvents were emitted affecting the photochemical ozone creation potential of the crane. Transports cause only a minor part of the emissions.

About 97% of the total amount of energy needed to manufacture and operate a forwarder crane is consumed at the use phase (Table 2). During its lifetime the crane will need 1050 MWh in form of diesel fuel (100 tons of diesel oil). At the same time the crane will handle about 680 000 m3 of wood. In energy terms this amount of wood equals to more than 2 750 000 MWh.

The production of steel was the second more significant stage in the life cycle of the crane. Steel is a central material in the crane comprising 98% of its mass. The process of making steel uses significant amounts of energy (6 kWh/ kg steel produced) and significant amounts of carbon dioxide are released (558 g/kg steel).

Table 2 Energy use and environmental impact per 1000 m3 vob for the life cycle stages of the forwarder crane.

Life cycle stages Energy use and environmental impacts

kWh g CO2 eq. g SO2 eq. g ethene eq. g O2 eq.

Raw material 29 9 610 20 0.1 83

2 % 2 % 0.5 % 0.04 % 0.3 %

Fabrication 17 1200 3 4.5 10

1 % 0.2 % 0.09 % 2 % 0.03 %

Transport 0.15 140 1 0.1 8

0.01 % 0.03 % 0.03 % 0.04 % 0.03 %

Use 1517 530 000 3 450 215 28 00097 % 97.8 % 99.3 % 97.9 % 99.6 %

TOTAL 1 565 540 000 3 474 220 28 100100% 100% 100% 100% 100%


53

Conclusions Using LCA methodology priorities for environmental improvements can be set. In the case of the forwarder crane it is apparent that fuel consumption and emissions to air during the operation phase should be targeted for reduction. This can be achieved by shifting to modern engines that consume less fuel or emit fewer pollutants. By using a low sulphur diesel oil, the SOx emissions and thus the acidification potential of the system can be reduced substantially. At the manufacturing stage solvent emissions could be reduced by substituting solvent borne by water borne paints.

Through an accredited certification body an external review of the study was conducted. It was confirmed that the data and the declaration is in confor-mance with the requirements for a certified Environmental Product Decla-ration. The final document, the first to be issued world-wide for a forest machine component, was approved on November 2001 and the EPD certi-fication was issued shortly after the approval.

Acknowledgements The project has been very successful in achieving its aims. The author and project leader acknowledges the valuable contribution of the project team (Jan Kärnestad, Cranab and Nina Åkerback, SYH) and of the members of the Reference Group (Stina Frejman, SYH; Ola Kåren, Holmen; Kjell Rönnholm, Cranab; Ulf Wiklund Tyréns and Iwan Wästerlund, SLU) who gave generously of their expertise at key points during the project. Further, the involvement of many of the employees at Cranab and suppliers is acknowledged and greatly appreciated. The study was partly financed by the Division of Forest Technology, Swedish University of Agricultural Sciences.

References Athanassiadis, D. 2000. Resource consumption and emissions induced by logging

machinery in a life cycle perspective. Doctoral Thesis. Swedish University of Agricultural Sciences, Department of Silviculture, Section of Forest Technology. Acta Universitatis Agriculturae Sueciae. Silvestria 143.

International Standards Organization, 1999. ISO 14041: Environmental management - LCA - Goal and scope definition and inventory analysis

Swedish Environmental Management Council 2000. Requirements for Environmental Product Declarations - EPD an application of ISO TR 14025 Type III environmental declarations. MSR 1999:2.


54

Profile Dimitris Athanassiadis Independent researcher, consultant and lecturer

Leading expert in Life Cycle Assessment, ISO 14000, Environmental Product Declarations, Forest Technology, Environmentally Conscious Manufacturing. Born: 11 April 1966 Nationality: Greek, Swedish Ph.D in Forestry, Swedish University of Agricultural Sciences, Umeå

(2000). Thesis title: "Resource consumption and emissions induced by logging machinery in a life cycle perspective".

Postdoctoral fellow at Forest Engineering Research Institute of Canada, Eastern Division (2002/2003).

Swedish representative in the Management Committee of Cost Action 530 ”Sustainable Materials Technology - Life Cycle Inventories for Environmentally Conscious Manufacturing Processes”. (2001/2006).

Athanassiadis Dimitris Life Cycle Assessment Specialist Ph.D in Forestry Gnejsvagen 47 BV 90740 Umeå Sweden Tel: +46 (0)90 719824


55

The challenges in developing the wepp cumulative effects model

William J. Elliot and Randy B. Foltz Project Leader and Research Engineer

Rocky Mountain Research Station Moscow, Idaho, USA

Summary Many of the forests in the U.S. and elsewhere in the world are source areas for water. The quantity and quality of this water are major public concerns. In a forested watershed, any road segment, harvesting operation, or other manage-ment activity can adversely impact forest streams. These disturbances are distributed in both time and space. The disturbance in the first year may have minimal impact on the hydrologic integrity of the watershed, but if the distur-bance in the following year is added to the first, and the disturbance in year 3 added to those in years 1 and 2, the net effect may be detrimental to the beneficial uses of the stream. A model to address this cumulative impact is sometimes referred to as a cumulative effects model. This paper presents the application of the GeoWEPP Geographic Information System (GIS) tool to evaluating cumulative effects in forests due to fuel management activities. An example is given to demonstrate the utility and limitations of the current tool.

Keywords: Watershed Analysis, Soil Erosion.

Introduction Forests provide society with numerous resources including fiber, food, recrea-tion, and water. Activities associated with obtaining some of these resources may adversely affect others. One conflict, in particular, is that any disturbance associated with obtaining fiber or food, and many recreational activities can adversely impact forest water quality.

A single disturbance in a given year is seldom a problem. Forest watersheds are able to recover within a few years from most single disturbance events, including disturbances as extreme as wildfire. As more disturbances are added during a year, and additional disturbances in the years that follow, the forest is less likely to recover to an undisturbed condition. The cumulative effects of numerous disturbances over a number of years must be considered to be able to manage forests for multiple uses.

The Water Erosion Prediction Project (WEPP) (Flanagan and Livingston, 1995) was developed by a number of United States Department of Agriculture Research and management agencies. Scientist at the Rocky Mountain Research Station and elsewhere parameterised the model for forests (Elliot and Hall,


56

1997). The WEPP model was released with both a “hillslope” and a “water-shed” version. Developing topographic input files for the watershed version was not easily achieved until in 2001, when a Geographic Information System (GIS) tool was developed to assist in spatial analysis and visualization of ero-sion distribution (Renschler et al., 2002).

Application of GeoWEPP to Watershed Analysis To evaluate the suitability of the GeoWEPP tool, an example study was carried out on a 1490 ha watershed about 25 km north of Moscow, Idaho (Figure 1).

Figure 1. Output from year 12 of simulations. Areas near outlet have recovered, and areas near the center of the watershed are recovering from forest operations and prescribed fire. The darker the area, the greater the erosion rate. Predicted erosion rate in the white is zero, the lighter shade, 0.1, the medium shade, 0.3, and the dark shade 1.4 t/ha.

The GeoWEPP tool divided the watershed into 33 hillslopes, and 13 channel segments. The watershed is currently under consideration for significant fuel reduction activities, including small diameter logging in year one, prescribed fire in year 2, and recovery of hydrologic stability and vegetative cover during the next five years. Table 1 shows the sequence of vegetation and soil proper-ties necessary to sequentially describe these disturbances and recovery years.

Table 1. WEPP vegetation and soil template values used for the analysis, assuming a silt loam soil.

Year Vegetation Hydraulic Conductivity (mm/h)

Rill Erodibility (s/m)

1 Established Forest 28 0.0004 2 Harvest: 80 percent cover, Young forest 23 0.0004 3 Burn: 80 percent cover, Low severity fire 13 0.0005 4 90 percent cover, Short grass 11 0.0004 5 95 percent cover, Tall grass 23 0.0004 6 95 percent cover Young forest 23 0.0004 7 100 percent cover Young forest 23 0.0004 8 Established Forest 28 0.0004

To demonstrate the application of GeoWEPP, each year a hillslope was selected, starting with hillslopes at the bottom of the watershed, to initiate the fuel reduction sequence. We assumed that all other hillslopes were covered in forest at the start of the simulations. Figure 2 shows the sediment yields for


57

the first 12 years of analysis, for both the disturbed hillslopes and the road network. Note that the first year assumed that all hillslopes were undisturbed, and the majority of the soil erosion was from the road. During the years of this example, the sediment yields varied between 40 and 90 tonnes, depending on the size and location of the disturbed hillslopes.

0

20

40

60

80

100

0 2 4 6 8 10 12

Year

Perc

ent o

f Wat

ersh

ed o

r to

nnes

sed

imen

t

% Foresttonnes Sediment Yieldtonnes Road Sediment

Figure 2. Percent of watershed In forest during the first 12 years of fuel reduction in watershed, and the associated sediment yields from roads and fuel management activities.

To consider the sediment from roads, sediment delivery was modelled assuming a road erosion rate of 1.33 t/km on roads with heavy traffic, and 0.67 t/km for roads with light traffic. These values were estimated with the WEPP model for multiple 60-m long road segments with a gradients of 4 percent, distances of 20 m between the road and the stream, and with buffers covered in forest. The rill erodibility value was reduced from 0.0003 s/m for the road with traffic to 0.000075 s/m for the road with low traffic, to reflect the observed surface armouring on roads without traffic (Foltz, 1998). It is apparent from figure 2 that the sediment from the road accounts for about a fourth of the sediment generated from human disturbances during active years, and 96 percent of the sediment in the absence of disturbances. The road sediment delivery values are approximate estimates in this study, as a detailed road map was not available. The relative importance of roads in the analysis, however is unlikely to change with greater detail.

These sediment yield rates need to be compared to the expected sediment yield from natural disturbances. When the entire watershed was described as wild-fire, the predicted sediment yield was 4832 tonnes in the year of the fire. If the frequency of fire in this area is assumed to be about 48 years, then the average annual sediment generated in the year following the wildfire averages about 100 tonnes per year. If fuel management operations reduce the likelyhood of fire, or the severity of the fire, as has been observed in recent studies, then the average annual sediment production due to the operations is less than sediment from wildfire.

To complete the analysis, some users may wish to add in sediment from land-slides. McClelland et al. 1997, found that typical sediment yields averaged over the 20 year return period associated with such events was around 10 t/ha.


58

Operations are unlikely to decrease this value, but a more dense road network could increase it.

Currently, the WEPP model only predicts surface runoff. Observations in many steep forest watersheds have shown that over 99 percent of all runoff is subsurface flow. Work is ongoing to incorporate subsurface flow into the WEPP model (Wu et al. 2000).

In summary, we have presented the application of the new GeoWEPP spatial analysis tool to cumulative watershed effects analysis. At this time, the tool is run for each year of disturbance. If desired, users can then add the sediment impact due to roads, wildfire, or landslides.

References Elliot, W. J. and D. E. Hall. 1997. Water erosion prediction project (WEPP) forest

applications. Ogden, UT:U.S. Department of Agriculture, Forest Service, Intermountain Research Station. Gen. Tech. Rep. INT-GTR-365: 11 p.

Flanagan, D. C. and S. J. Livingston. 1995. WEPP User Summary, USDA-Water Erosion Prediction Project (WEPP). W. Lafayette, IN: USDA-ARS National Soil Erosion Research Laboratory. 123 p.

Foltz, R. B. 1998. Traffic and no-traffic on an aggregate surfaced road: Sediment production differences. Proceedings of the seminar on environmentally sound forest roads and wood transport. Sunnia, Romania, 17-22 June, 1996. Rome: FAO.

McClelland, D. E., R. B. Foltz, W. D. Wilson, T. Cundy, R. Heinemann, J. Saurbier, and R. Schuster. 1997. Assessment of the 1995 and 1996 floods and landslides on the Clearwater National Forest. Part I: Landslide assessment. Missoula, MT: USDA Forest Service, Region 1. 52 p

Renschler, C. S., D. C. Flanagan, B. A. Engel, J. R. Frankenberger, T. A. Cochrane and R. C. Vining. 2002. GeoWEPP - The Geo-spatial interface for the Water Erosion Prediction Project (WEPP). Online at: < http://www.geog.buffalo.edu/~rensch/geowepp > Accessed 2002 December 4.

Wu, J. Q., A C. Xu,. and W. J. Elliot. 2000. Adapting WEPP for forest watershed erosion modeling. Paper No. 002069. Presented at the 2000 International ASAE Meeting, July 9-12, Milwaukee, WI. St. Joseph, MI: ASAE. 9p.


59

Profile Randy B. Foltz is a research engineer for the United States Forest Service located at the Intermountain Research Station in Moscow, Idaho. He has worked in the area of erosion from forest roads for 15 years and is an author or co-author of over 40 papers in this area. Dr. Foltz has a PhD in Civil Engineering from the University of Idaho, a Master’s degree in Civil Engi-neering from New Mexico State University, and a Bachelor’s degree in Chemical Engineering from New Mexico State University.

William J. Elliot Randy B. FoltzRocky Mountain Research Station 1221 South Main Moscow, Idaho USA

Research Engineer RMRS, Moscow 1221 S. Main St. Moscow, ID 83843 [email protected]

208 883 2338 (work) 208-301-4511 (mobil)

[email protected]


60

The use of the GIS into the forest fire prediction

The simulation model Ahmed Saidi, A. Missoumi,

Centre National des Techniques Spatiales, Laboratoire de Géomatique, Algeria Summary

The mastery of strategies of forests fire-fighting passes compulsorily by deep knowledge of forest fire phenomenon. One of the efficient means of appre-hension of the forest fire, is to dispose of a tool that is capable to inform us about the behavior of fire before its apparition according to the given climatic conditions. In this context, the simulation remains an effective tool for the prediction of the fire behavior. It permits to determine with a relative confi-dence degree, susceptible zones to be ravaged by the fire during a determined period.

The objective of this work is the study of mechanisms of forest fire progress-sion by the elaboration of an automatic tool capable to pattern suitably a fire forest, its parameters, its propagation and its behavior in a given region. Through out this study, it will make conspicuous the considerable property (perhaps unavoidable) the Geographical information Systems (GIS), in combination with techniques of simulation in the apprehension of a proble-matical " fires forest " [DAG 94]. This resides in the power of the GIS to modelling all phenomenon presenting a geographical character.

The interest that presents such a survey for operators in charge of the manage-ment of a forest fire (Fireman department, services of forests, local collective-ties, etc.) is double. It permits to define to the long term a homogeneous and coherent politics of forest fire prevention, whereas the system permits to verify the adequacy of amenities and presents infrastructures of wrestling against the fire with the reality of a disaster [DAG 97]. It also allows to determine, in the perspective of the beginning of a fire, means to put in work for a coordination of intervention teams and a strategy that remains efficient, of wrestling against the fire progression

Keywords : Fire forest – Propagation - GIS – Simulation – Modeling. Problematic

A lot of countries confronted to problems of forest fire seized the interest to resort to modern techniques of science to control and to master the forest fire that is source of permanent danger for the nature, the environment and the man's security. Countries as Canada, USA, France, Germany, Austria, etc., have started during the years 70 the study and the development of relative computer systems to a thematic “forest fires”.


61

In the panoply of developed tools, models of propagation represent an appreciable part. Indeed, to fight a forest fire efficiently, a fire passes cum-pulsorily by the understanding of mechanisms governing its propagation. Among techniques used for the propagation phenomenon study, the simulation holds a preponderant place.

Parameters of forest fire definition

The forest fire integrates the ambient air and draw its fuel in the plantable setting. Wind constitutes its principal vector. The shape of the land where it appears, contributes to its development and extension.

In this context, the main factors intervening in the evolution of a forest fire, are according to SHERLIS [SHE96] and MISSOUMI [MIS97] strength and the direction of wind, the degree of inflammability of the plantable setting, the importance and the orientation of the slope and the starting point of fire.

Presentation of the study Our study will note the contribution of a Geographical Information system (GIS) in the development of a model of forest fire simulation. The recourse to the GIS is justified by the geographical nature of the data relating to forest fire. Moreover, the tools of the graph theory and mathematical models necessary for simulation, are usual components in the GIS tools.

GIS and modeling forest fires The factors speed and direction of the wind, are extracted from weather data and the combustibility index of vegetable cover results from the remote sen-sing. Once these parameters injected into the model, the zone obtained by simulation looks like the shape of a polygon which expresses in our view a better approximation of the zone to be devastated by the forest fire. The GIS make it possible to develop automatic processes, likely to generate the zones of propagation of fire according to the developed model.

In the context of our study, it is a question to represent the propagation of the forest fire by a GIS model. the model of simulation introduced represents a modeling in a GIS, of a uniform movement disturbed by parameters decelera-ting its speed. The movement starts at the starting point of fire and the dis-tance covered represents the ray of the circle. According to a preset step by an adequate sampling, the same operation is repeated in a circular sweeping of space. This results in a different distance for each ray (influence of the para-meters). At the end of sweeping we obtain a polygonal form whose vertices are the extreme points of the rays. This form predicts in our view the zone devastated by fire under similar conditions of wind and climate.


62

0FF

x

3,1

321

1&1

),,(**

)sin()cos(

iii

MtIcDvMtIcDvfct

FtVfRRYdyRXd Xd, Yd : fire starting point.

Vf : Wind speed. T : Propagation time. Dv : Wind direction. Ic : Index of combustibility. Mt : Digital Terrain Model. : Sampling Angle. : contribution coefficient.

In the presented model, the propagation is defined like a continuous move-ment on an interval of established time. This movement will be disturbed by the action of the various parameters taken into account the development of a forest fire. Experimentation The model was tested on station DEC ALPHA STATION via software GIS Arc/Info 7. procedures of simulation were developed on PC by the software Arc/Info PC.

Fire extent after 1 hour.

Fire extent after 4h.

Wind variation 2 (weak wind) Direction N NE.

MF IcDv t*25.0*25.0*5.0



Wind variation 3 (strong wind) Direction N NE.

MF IcDv t*2.0*2.0*6.0


63

Experimentation The model was tested on station DEC ALPHA STATION via software GIS Arc/Info 7. procedures of simulation were developed on PC by the software Arc/Info PC.

Conclusion

The introduction of the GIS makes it possible to deal with the problem of forest fire propagation with much rigour and better precision. However, the system of simulation developed here, is not released to answer a strategy of intervention in real time. Its precision is still very approximative. This is not due to the adopted approach, but to the limits of the current tools to modelling the intervening factors in the manifestation of the forest fire. Indeed, The natural elements that are the wind, and vegetable cover are much more com-plex to be represented by simple indices and values. Science does not provide at present time other tools that the wind speed and its direction to modeling an air movement on a given period, and the index of combustibility for the vegetable species in interaction with fire.

To claim to compute exactly the extent of fire through a simulation, it needs a knowledge of speed and direction of the wind over periods much shorter, even instantaneous. The important influence of this last factor on the fire propaga-tion phenomenon, can become aware of the extent of the error made in the prediction. Only a more rigorous mathematical model in its definition and modelling of the wind can mitigate the effect of disparity of the simulation model with the reality. The index of combustibility of a vegetable species is a measurement much too rigid to contribute in the model of propagation with-out too many errors.

This established fact shows the difficulty to modelling suitably the propagation of forest fire. Many systematisms which influence is considerable, are occulted in the current processes of simulation. This is a consequence of the inexistence of models governing their demonstration.

References [CAR96] P.CARREGA et J.L.WYBO, Vers une évaluation du risque d'incendie de forêt.

[DAG94] A.DAGORNE, SIG, télédétection aérospatiale et gestion des espaces sensibles aux feux et/ou parcourus par eux...ou l'utilisation de la cartographie.

[DAG96] A.DAGORNE, Application d'un SIG pour l'évaluation de la vulnérabilité au feu et la prévention.

[DAG97] A.DAGORNE et J.Y.OTTAVI, Des données à l'information, ou l'utilité d'un SIG.

[ISH96] R. INSAK, Y. STARP, J.F. HOLBY Predominance of factor wind in the development of natural disaster effects.

[KNU85] L.J. KNUTH A principles of statistics and econometry Vol 2.

[LEN98] M.LENCO et B.KIENTZ, Etude par télédétection de la simulation du déroulement du feu de forêt du massif de Sainte Victoire.


64

[MIS97] A. MISSOUMI Caractérisation des zones forestières à risque d'incendie à l'aide d'un Système d'Informations Géographiques.

[PAR90] A. PARENT Un système automatique pour quantifier le feu de forêt. Cas de la forêt de chicoutimi.

[SHE96] A.J. SHERLIS An overview on the Main factors in forest degradation. The fire forest contribution.

Profile I am a geomatic Expert, I am a young researcher in the laboratory of geomatic of the National Center of Space Techniques (CNTS Algeria), I am responsible for several projects in the GIS field and their applications, particularly in the environment, the forests, the networks, urban management, remote sensing, etc. I teach the SIG and the urban data bases to the engineers and the magisters at the CNTS. I published a several ten articles (11) and I took part in many international scientific events.

Currently I am responsible for a team of research for the development of tools of decision-making in the geomatic technics.

Dr Ahmed Saidi, CNTS bp 13 Arzew 31200 ALGERIATél : (213) 41 47 25 82, Fax : (213) 41 47 35 65Email : [email protected] [email protected]


65

A new grading standard for softwood based on optical

logscanning

– Results from a study on Scots pine Jacob Edlund

PhD student, Department of forest management and products, Swedish University of Agricultural Sciences, Box 7060, 750 07 Uppsala Sweden

Summary Log grading for the pricing of softwood in Sweden is based on visual assess-ment of log characteristics. The grading method relies on subjective judgments, made over a short period, for each log and could be considered time consu-ming. One method of simplifying log grading would be the use of diameter sorting equipment to obtain variables from each log that describes its shape, and use these variables to approximate the grade. The main objective of this study was to evaluate the possibilities of developing an automatic grading routine for pine saw logs using external geometry features describing log shape obtained from 3D log scanner. The study was based on a sample of 433 Scots pine logs taken randomly from 3 sawmills in the northern-, middle- and southern part of Sweden. Models for grading logs according to the current grading system were developed from the data set using linear discrimination analysis and the agreement between the true grading and the model was evalua-ted using a chance corrected coefficient, the simple kappa (k). Preliminary results indicate that automatic grading of logs according to the new grading system is comparative to the system of today both in terms of precision and repeatability.

Keywords: External shape, log grade, log scanners, measure, saw logs.

Introduction Log grading for the pricing of softwood in Sweden is based on visual assess-ment of log characteristics, where the graders manually grade the log into five or six grades. The current grading rules (Anon, 1999a) are standardized for the whole of Sweden and grading is performed by neutral log grading associations organized under VMR (the timber measurement council). The grading method relies on subjective judgments, made over a short period and could be conside-red intricate, time consuming and monotonous. One method of simplifying log grading would be the use of diameter sorting equipment to obtain variables from each log that describes its shape, and use these variables to approximate the grade.

Properties that influence log quality and log grade can be divided into those that are not related to stem form and those that are, examples of the latter are external geometry features like taper and bumpiness which can be used to pre-dict log type and grade. In the current Swedish grading system, that is used for


66

pricing, Grade 1 and Grade 3 logs can be sorted using butt taper whereas Grade 2 logs can be sorted using top taper. Unevenness is also indirectly rela-ted to the current grading rules as log surface unevenness indicates big knots that can cause down-grading to poorer grades or indicates logs at the top of the stem; poor grades are 5 and cull. Poor grades has often down-grading causes such as, rot, ring width, compression wood and such properties that are not related to outer shape.

Earlier attempts to sort logs using diameter sorting equipment have been con-ducted (Grace, (1994); Nylinder, (1990); Lundgren, (2000); and Jäppinen, (2000). Lundgren and Jäppinen (1998) determined that variables calculated from data from a 3D log scanner could be used to sort saw logs into two classes for example suitable or not for glue laminated lumber and top or butt log.

The main objective of this study was to evaluate the possibilities of deve-loping an automatic grading routine for pine saw logs using external geo-metry features describing log shape obtained from 3D log scanner and grades according to the current Swedish grading standard for pricing.

Materials and method Dataset The study was based on a sample of 433 Scots pine logs taken randomly from 3 sawmills in the northern-, middle- and southern part of Sweden. The goal was to collect 40 logs from each diameter class and each log grade were the diameter classes were small, medium and large. The grades were according to the Swedish Measurement Council and comprised 5 grades and cull. All logs were graded, numbered and the cause of down grading recorded by a grading inspector. The selected logs were scanned twice using the Rema 3D log-scanner. These data were used to calculate variables describing log shape.

Table 1. Number of logs in the training set by current grade and diameter class Top diameter under bark Mm

Grade diameter-class 1 2 3 4 5 Cull

(152-163) 17 38 39 52 31 4 181 (226-243) 28 5 40 42 30 13 158 (335-359) 11 0 30 30 16 7 94

grade 56 43 109 124 77 24 433

New grades Earlier studies has shown that Grades 1 and 3 have similar outer shape proper-ties and were therefore combined into one grade - Grade 1+3. Grade 2 and grade 4 differed from other grades with respect to outer shape and remained as grades. Logs that had been reduced in volume or downgraded to grade 5 be-cause of crook, compression wood or rot was set to a new grade, “second rate”. The grade “second rate” was not evaluated in this study. Though, if an automatic grading system were to be used, it would be necessary to have a system for manually or automatically downgrading logs or reducing the price

Comment [SLP1]: Page: 5 If I have understood correctly you have only focussed on two species of wood – pine and spruce, this was not so clear from the text

Comment [SLP2]: Page: 5 So VMR is the current grading system in Sweden? If so it might be an idea to use this abbreviation when you refer top the ’current’ system


67

for “second rate” logs. This could be done either per log or for a whole lot. according to figure 1.

Figure 1. Share of “second rate” timber is visually assessed either per log or per lot, the logs are then automatically graded.

Models Models for grading logs according to the current grading system were developed from the data set using the grade set by the inspector and the variables calculated from the raw data from the 3D log scanner. The SAS function for linear discrimination analysis was used to establish the linear functions for each grade (Anon., 1999b).

Analysis The agreement between the true grading and the model was evaluated, primary using the simple kappa (k) coefficient (Equation A). As kappa coefficient is chance corrected, it is more advantageous than the commonly used proportion of observed agreement (P0) which though also was calculated (Cohen, 1960). The equation shows that the kappa value is calculated by correcting the proportion of observed agreement (P0) with the chance factor (Pe)

eP1r

iiipP0

ePP0

r

iie ppP

[Equation A]

where: and , i

Results Models The standardized parameter estimates for the linear discriminate functions were calculated for each grade using the data set (table 2). The variables are described in a study by Lundgren (2000).

Grade 1+3 Grade 4 Grade 2

Alternative 1 Assessment per lot

Alternative 2Assessment per log

Rot Automatic grading

Crook

Second rate


68

Table 2. Standardized parameter estimates for a discriminant model explaining the probability for each grade and log. Variable f2(x) f1+3(x) f4+5(x) Butt taper -0.90 0.45 -0.15

Top taper/ Diam 0.99 -0.30 0.03

Top taper Uneven0

0.51 -0.29 0.11

Uneven4 0.47 -0.40 0.18 Butt taper/ Diam -0.80 0.43 -0.15

Mx taper -0.85 0.30 -0.05

Mndif2 0.29 0.13 -0.14

The accuracy of the automatic grading for all logs and the accuracy for diffe-rent diameter classes is shown in table 3. The simple kappa coefficient for the automatic grading of all logs was 0.39. The accuracy for automatic grading in-creased with an increase in diameter.

Table 3. Kappa values ( ), observed agreement (P0) and chance factor (Pe) for different diameter classes in automatic grading of pine logs . Top diameter under bark (mm)

Simple kappa

( )

P0 Pe Number of logs

All logs 0.39 0.65 0.43 408 (152-163) 0.35 0.59 0.37 170 (226-243) 0.35 0.67 0.49 150 (335-359) 0.44 0.73 0.52 88

Discussion Preliminary results indicate that automatic grading of logs according to the new grading system is comparative to the system of today both in terms of preci-sion and repeatability. The new grades distinguish between logs with different board quality comparatively to the system of today.

The aim of the grading is to give the sellers a fair price according to the quality of the delivery, log properties that cannot be detected by a logscanner e.g. rot and compression wood should therefore be taken into account in the grading system in an alternative way. Some sorts of manual assessment have to be made, preferably by visual assessment per log or of a carload.

Acknowledgements The Swedish Timber Measurement Council financed this work.

Comment [SLP3]: Page: 10 no footnote for this

Comment [SLP4]: Page: 10 see 42


69

References Anon, 1999a. Mätningsinstruktioner för rundvirkessortiment, VMR cirkulär Nr 1-99.

The timber measurement council, Märsta Sweden

Anon., 1999b. SAS Online Doc. SAS Institute Inc., Cary, NC, USA.

Cohen, J., 1960. A coefficient of agreemnent of nominal scales. Educational and Phychological Measurement, 70: 37-46.

Grace, L.A., 1994. Design and Evaluation of an Optical Scanner Based Log Grading and Sorting System for Scots pine (Pinus sylvestris L.) Sawlogs. Dissertation Thesis, Swedish University of Agricultural Sciences, Uppsala, 23 pp pp. ISBN 91-576-4848-4

Jäppinen, A., 2000. Automatic sorting of sawlogs by grade. Dissertation Thesis, The Swedish University of Agricultural Sciences, Uppsala ISSN 1401-6230

Jäppinen, A.&Lundgren, C., 1998. Grading of logs using external geometry features from 3D- and profile log scanners - a comparison. In: O. Lindgren, A. Grönlund and O. Hagman (Editors), 3rd IWSS. Luleå University of Technology, Skellefteå, Sweden, pp. 31-38. ISSN: 1402-1536

Lundgren, C., 2000. Predicting log type and knot size category using external log shape data from a 3D log scanner. Scan J of For Res, 15: 119-126.

Nylinder, M., 1990. Automatic grading of pine logs. Results from investigations at Rockhammars sawmill. 215, Department of Forest Products, Swedish University of Agricultural Sciences, Uppsala.ISSN 0348-4599.

Profile My name is Jacob Edlund I am a PhD student at the Institution of Forest Managements and Products at SLU in Uppsala, Sweden. I am working on with a project that evaluates the possibilities to automatize the grading for saw logs using external geometry features. The work is done in close cooperation with the Swedish Timber Measurement Council in a large national project. My formal education is Master of Science in Forestry with emphasis on economics and marketing. I have during my PhD studies studied and practiced a great deal of statistics and modeling.

Jacob Edlund Department of forest management and products, Swedish University of Agricultural Sciences, Box 7060, 750 07 Uppsala Sweden


70

Impact of harvesting age on end-product quality in Jack Pine

(Pinus banksiana Lamb.) Isabelle Duchesne, Gilles Chauret and S.Y. Zhang

Research Scientist, Forintek Canada Corp., Eastern Laboratory Canada

Summary In many countries including Canada, there is a current trend to allocate more forests for conservation, eco-tourism or non-harvesting activities. This will reduce the forest areas available for wood production. There has also been an increased demand for Canadian wood products which puts a high pressure on Canadian forests to produce more wood. To sustain wood supply, silviculture will have to be intensified in Canada to counteract the effects of extensive forest harvesting and extremely slow tree growth rates. Intensive forest management practices are only starting to be applied on a larger scale. The impact of some forest management decisions have already been determined for a few Canadian commercial species (e.g. black spruce, white spruce). To ensure the viability of the wood industry it is important to understand the impact of forest man-agement not only on growth and yield, but also on end-product quality and value. The present study evaluates the impact of harvesting age on wood characteristics and product quality to determine the rational harve-sting age for natural jack pine stands.

Three jack pine stands aged 50, 73 and 90 years were selected in the boreal forest of Northern Ontario (Timmins region) where jack pine is a major commercial species. The natural stands were all established after forest fires on well-drained sandy soil and were located within a 5 km distance from each other. A total of 147 trees were collected. For each stand, 6 trees per DBH class were selected to cover all DBH classes in 2-cm interval (from 10 to 30 cm whenever possible). For each sample tree, major tree characteristics were measured: 1) total tree height, tree height up to 7 cm diameter top, tree height up to 9.01 cm diameter top (10 cm DBH class), 2) DBH and stem diameter from the stump to the top at 1-m interval, 3) live crown width and length and 4) average dia-meter of the 5 largest branches. Based on these measurements, other tree characteristics were calculated: 1) stem volume, 2) stem taper, and 3) length of the log below live crown. Each sample tree was bucked to 2.5-m-long logs for lumber conversion. From the base of the stem and from the top of each log, a 5-cm-thick disc was removed for the evalua-tion of wood characteristics. Lumber conversion was carried out in a way which allows us to keep track of the provenance of each piece of lumber (over 800 logs were sawn). Each piece of lumber was visually graded


71

after drying and planing in accordance with the National Lumber Gra-ding Association (NLGA) rules. Detailed results on lumber strength and stiffness (modulus of rupture (MOR) and modulus of elasticity (MOE)) and wood properties (e.g. annual growth ring width, basic density, etc.) as related to stand age will be presented at the poster session of the conference.

Table 1. Stand characteristics for three jack pine stands studied. The stand density decreased with age while tree mortality tended to increase. Stand characteristics Age 50 73 90 Stand density* (trees/ha)

1 550

1 300 1 075

Tree mortality (%) 16–22

10–20 33–34

Other species** (%) 2-15

6–10 0-2

*Based on random plots, ** Black spruce, white spruce, balsam fir and white birch.

50 yr-old stand

05

101520253035

8 10 12 14 16 18 20 22 24 26 28 30 32

DBH class (cm)

Freq

uenc

y (%

)

other speciesdead jack pine treesliving jack pine trees

73 yr-old stand

05

101520253035

8 10 12 14 16 18 20 22 24 26 28 30 32

DBH class (cm)

Freq

uenc

y (%

)


90 yr-old stand

Freq

uenc

y (%

)

05

101520253035

8 10 12 14 16 18 20 22 24 26 28 30 32

DBH class (cm)


Figure 1. Diameter class distribution of the three sample jack pine stands located in the boreal forest of Northern Ontario


72

Figure 2. Natural 73-year-old jack pine stand selec-ted for the study. Trees were bucked into 2.5 m-long logs, and wood discs (third picture taken at the 90 year-old stand) were sawn at the top of each log for analysis of wood characteristics.

Figure 3. Lumber conversion of over 800 logs was conducted in a way which allowed us to keep track of each piece of lumber produ-ced. Each piece of nominal dimensions 2x3”, 2x4” and 2x6” was further tested for its mechanical properties (MOR/MOE).


73

Profile Isabelle Duchesne obtained her Bachelor’s Degree in Wood Science from Laval University, Quebec, in December 1991. Thereafter, she worked as quality controller and production foreman in a large sawmill in Northern Quebec. In 1994, she moved to Uppsala, Sweden. She wor-ked as a research assistant on different wood and fibre quality projects at the Swedish University of Agricultural Sciences. Between 1997 and 2001, she completed licentiate and doctoral studies on the “Surface ultrastruc-ture of kraft pulp fibres” within the frame of the Wood Ultrastructure Research Centre located at SLU, Uppsala. Since April 2002, she has been working as research scientist at the Resource Assessment and Utilization Department of Forintek Canada Corp - The National Wood Products Research Institute of Canada - and involved in a number of projects dealing with the impact of silviculture on the quality of fibre-based products.

Isabelle Duchesne, Ph.D. Forintek Canada Corp. 319, rue Franquet Sainte-Foy, Québec G1P 4R4 Canada

1-418-659-2647 (work) [email protected]


74

Life cycle carbon dioxide (lcco2) analysis of residual forest biomass as an alternative energy resource in

Japan

Takuyuki Yoshioka, Postdoctoral Researcher

Kazuhiro Aruga, Assistant Professor

Toshio Nitami, Associate Professor

Hiroshi Kobayashi, Professor The University of Tokyo

Tokyo, Japan

Summary The amount of carbon dioxide emission of residual forest biomass as an alternative energy resource in Japan over its entire life cycle, i.e., the life cycle carbon dioxide (LCCO2), has been analyzed using the life cycle inventory (LCI) analysis method. Furthermore, the reduction in the amount of carbon dioxide emission that would result from replacing coal-fired generation systems with a biomass-fired generation system in Japan has been estimated. Coal-fired gene-ration systems have a higher environmental impact than any other power-generation technology. Fuel consumption by forestry machines was measured in field experiments on the harvesting of residual forest biomass in Japanese forestry operating sites. In addition, the carbon dioxide emission of the total system including materials, construction, repair and maintenance of forestry machines as well as the energy-conversion plant was evaluated. The carbon dioxide emission per MWhe (e: electricity) of the biomass-fired generation system (residual forest biomass is supposed to be fired) was calculated to be 61.8 kgCO2/MWhe, while that of the coal-fired generation system in Japan is 960 kgCO2/MWhe. Therefore, the reduction in the amount of carbon dioxide emission that would result from the power generation of as much as 3.0 Tg/y of residual forest biomass (on a dry-weight basis) was estimated to be 1.66 TgCO2/y. This amount corresponds to 0.146% of the national carbon dioxide emission estimated for 2001, i.e., 1.14 PgCO2/y. It was clarified that it is possible for Japan to reduce its domestic carbon dioxide emission by utilizing residual forest biomass as an alternative energy resource.

Keywords: Carbon dioxide (CO2) emission factor, feasibility study in Japan, gross energy analysis, life cycle inventory (LCI) analysis, residual forest biomass.


75

Introduction Nowadays, biomass as a renewable and carbon-neutral energy resource is becoming the focus of attention because of the increasing danger of global warming. Among various biomass resources, woody biomass is particularly attracting a great deal of attention in Japan. This is due to its abundance as well as the fact that the energy utilization of woody biomass is expected to contribute in two ways: the revitalization of forestry and forest products industries, which have been depressed for a long time, and the maintenance of the public benefit functions of artificial forests which are behind in tending. Many studies have been conducted on some energy-conversion techniques of woody biomass, which are already at the level of practical use. However, few studies have been carried out in Japan on the development of harvesting and transporting technologies. In order to promote the introduction and diffusion of bioenergy utilization, it is necessary to establish a low-cost harvesting and transporting system for woody biomass as quickly as possible.

Concerning the feasibility of a harvesting and transporting system for logging residues as residual forest biomass in Japan, some basic studies were conducted by the authors (Yoshioka et al. 2000, 2002, and in print) and the following conclusions were obtained. (1) It is not favorable from a standpoint of the harvesting cost. (2) However, there is no specific problem from the point of view of the energy input and output. (3) Furthermore, it is possible for Japan to reduce its domestic carbon dioxide emission by utilizing residual forest biomass as an alternative energy resource. In these studies, however, only the fuel consumption of each forestry machine was evaluated; that of the energy production system was not. In order to analyze energy input and output, for example, it is necessary to consider the energy required to manufacture machines and construct an energy-conversion plant in addition to the fuel consumption of machines as energy input.

In this study, to evaluate the feasibility of a harvesting and transporting system for residual forest biomass as an energy production system, the gross energy analysis and the carbon dioxide emission factor of residual forest biomass as an alternative energy resource in Japan over its entire life cycle were calculated. A detailed analysis was performed using the life cycle assessment (LCA) method, i.e., the life cycle inventory (LCI) analysis.

Life Cycle Inventory (LCI) Analysis Definition of the System Boundary The system was constructed using state-of-the-art technology and infrastruc-ture available in Japan. Namely, a power generation system that supplied energy by utilizing logging residues as fuel was assumed. The net power output and the thermal efficiency of the biomass-fired generation plant were supposed to be 3 MW and 12%, respectively. The system boundary is shown in Figure 1.

2nd rs Forest Engineering Conference Poste

76

System boundary

EquipmentOperation

EquipmentOperation

EquipmentOperation

Energy input Energy output

Manual felling (chain saw)

Processor limbing and bucking

Whole-tree skidding/yarding

Biomass-fired generation system

Power generation

Chipper comminuting

Forwarder hauling

Drying in a forest(Water content is reduced to 50%.)

Equipment

Operation

CO2

CO2

CO2

CO2

Electricity

Truck transporting

Drying in a stockyard(Water content is reduced to 15%.)

Figure 1. Definition of the system boundary, which is outlined by a broken line in the figure.

Gross Energy Analysis The “Equipment” energy, which is needed to manufacture machines and construct a plant, and the “Operation” energy, which is needed to operate machines and a plant over their entire life cycle, of all the processes inside the system boundary were estimated based on field experiments and existing studies. Then, the energy payback time and the gross energy analysis were calculated considering the energy input and output of the system.

As a result, with regard to the energy input, the amount of the energy required for operating the system (82% of the total) was fairly larger than that of the energy required by the pieces of equipment that made up the system (18%). Moreover, concerning the “Operation” energy, the processes of chipper com-minuting and truck transporting accounted for high rates of consumption, so it was ascertained that it is essential for the system to improve the operational efficiency of both processes.

The energy payback time was 1.09 year, showing that residual forest biomass was superior to other renewable energy resources, e.g., wind (1.99 year) or solar (10.00 year). In addition, the gross energy analysis was 5.69, indicating that the system examined in this study could be feasible as an energy production system.

Calculation of the Carbon Dioxide (CO2) Emission Factor The carbon dioxide emission factor of the system was calculated on the basis of the gross energy analysis. The carbon dioxide emission per MWhe was determined to be 61.8 kgCO2/MWhe, while that of the coal-fired generation


77

system in Japan is 960 kgCO2/MWhe. Therefore, the reduction in the amount of carbon dioxide emission that would result from power generation of as much as 3.0 Tg/y of residual forest biomass was estimated to be 1.66 TgCO2/y. This amount corresponds to 0.146% of the national carbon dioxide emission estimated for 2001, i.e., 1.14 PgCO2/y. It was clarified that it is possible for Japan to reduce its domestic carbon dioxide emission by utilizing residual forest biomass as an alternative energy resource.

Sensitivity Analysis Using sensitivity analysis, it was shown that the energy-conversion efficiency of a plant was the parameter which had the greatest influence on the results of this study. It was also suggested that small-scale and efficient energy-conver-sion technology of biomass should be developed.

Conclusion The reduction in the amount of carbon dioxide emission that would result from power generation of residual forest biomass was estimated. The result is not necessarily satisfactory, for example, when compared to the standards set by the Kyoto Protocol. This study, however, took only logging residues into account; therefore, it is essential to analyze the life cycle carbon dioxide emis-sions of thinned woods (5.0 Tg/y) and broad-leaved trees (9.0 Tg/y) according to the standards of the Kyoto Protocol.

References Yoshioka, T., Aruga, K., Iwaoka, M., Nitami, T., Sakai, H., and Kobayashi, H. in print.

Cost and carbon dioxide (CO2) effectiveness of fossil energy substitution with residual forest biomass in Japan. In Proceedings of the International Seminar on New Roles of Plantation Forestry Requiring Appropriate Tending and Harvesting Operations (September 29-October 5, 2002, Tokyo, Japan).

Yoshioka, T., Aruga, K., Sakai, H., Kobayashi, H., and Nitami, T. 2002. Cost, energy, and carbon dioxide (CO2) effectiveness of a harvesting and transporting system for residual forest biomass. Journal of Forest Research 7: 157-163.

Yoshioka, T., Iwaoka, M., Sakai, H., and Kobayashi, H. 2000. Feasibility of a harvesting system for logging residues as unutilized forest biomass. Journal of Forest Research 5: 59-65.


78

Profile Dr. Takuyuki Yoshioka Dr. Takuyuki Yoshioka is a Postdoctoral Researcher of the Laboratory of Forest Utilization (synonymous with “Forest Engineering” in Japan) of the University of Tokyo and a Research Fellow of the Japan Society for the Promotion of Science. He is now working under Professor Dr. Hiroshi Kobayashi on harvesting technology for logging residues as residual forest biomass in Japan, a country with steep topography. His doctoral thesis discussed the feasibility of a harvesting and transporting system for residual forest biomass in Japan. Dr. Yoshioka is currently pursuing a postdoctoral fellowship abroad in order to intensify his research in the field.

Kazuhiro Aruga Kazuhiro Aruga was awarded a Ph.D. in Agriculture by The University of Tokyo in March 1999 and has been an assistant professor at The University of Tokyo 's Graduate School of Agricultural and Life Sciences, Division of Forest Sciences since June 1999. He has carried out research into the impact of logging operations on forest environments. He is especially interested in soil compaction and tree root damages caused by soil compaction and has analyzed the interaction between forestry machines and soil by applying both numerical and experimental methods. This numerical method has a potential to be applied to any kinds of forestry machines.

Takuyuki Yoshioka, Ph. D. (JSPS Research Fellow) Laboratory of Forest Utilization Graduate School of Agricultural and Life Sciences The University of Tokyo 1-1-1 Yayoi, Bunkyo-ku, Tokyo 113-8657, Japan +81-3-5841-5205 +81-3-5841-7553 (fax)

[email protected]

Kazuhiro Aruga, Assistant Prof. Laboratory of Forest Utilization Graduate School of Agricultural and Life Sciences The University of Tokyo 1-1-1 Yayoi, Bunkyo-ku, Tokyo 113-8657, Japan +81-3-5841-5205 +81-3-5841-7553 (fax)

[email protected]


79

Ergo wood Ergo efficient mechanised logging operations

Sten Gellerstedt and Jerry Johansson Swedish University of Agricultural Sciences

www.spm.slu.se/ergowood/index.htm

Summary A three-year project funded by the European Commission has started with the intention to develop guidelines on ergonomic matters for European users, buyers and manufacturers of forest machines. This will encourage the development of safe and efficient forest machines, which are easy to use and maintain, as well as the improvement of the sustaina-bility in human resources. The project also involves sharing of good examples of work-crew building, work-shift scheduling, job-rotation and work enlargement in logging operation. Different ways of organising forest work will be investigated and assessed. The measured effects (output) will be presented in terms of economic, social and health out-put. Reliable measuring methods will be developed. This will make it easier to understand the benefits of ergonomic investments. Finally, the project also will contribute to make forest work attractive to young people.

Objectives Give the European logging industry a better competitiveness through;

Improvement of the sustainability in human resources.

Development of the organisation of logging operations.

Development of safe and efficient forest machines, which are easy to use and maintain.

Find reliable measuring methods for economical, social and health output will be developed.


80

Expected Achievements This project will develop, publish and initiate implementation of European recommendations for ergo efficient mechanised logging operations. A plan of implementa-tion includes e.g., regional seminars and examples of modules in education and training. The project focuses on industry groups with which participating part-ners already have links.

The output of the project, to be ready in fall 2005, includes:

Recommendations and experiences of different work organisations throughout the EU logging business, emphasising the process to achieve job rotation, work enlargement, participation etc.

Technical ergonomic guidelines for forest machines. These are deve-loped through structured and detailed discussions with all parts con-cerned in the participating regions, on ergonomic requirements for forest machine design, selection, operation and maintenance.

Measuring methods for the ergonomic guidelines for forest machines.

Standardised methods for the investigation of machine operators’ safety, health and social situation as well as for assessing work organisation.

A documentation of the machine operators’ safety, health and social situation establishing starting points which later can be followed up of other projects.

Work environmental monitoring systems with cost/benefit analysis methods.

Organisation Swedish University of Agricultural Sciences (SLU) Sten Gellerstedt, Scientific co-ordinator Dep. of Forest Products and Markets Jerry Johansson, adm. co-ordinator Box 7060 Phone: +46 18 673818 SE 750 07 Uppsala Fax: +46 18 673800 Sweden E-mail: [email protected] E-mail: [email protected] National Institute for Working Life/West (NIWL) Jørgen Winkel Box 8850 Phone: +46 31501600 SE 402 72 Göteborg Fax: +46 31 501610 Sweden E-mail: [email protected] Association Forêt Cellulose (AFOCEL) Maryse Bigot Domaine de l'Etançon Phone: +33 (0)1 60 67 02 34 77370 NANGIS Fax: +33 (0)1 60 67 00 27 France E-mail: maryse.bigot@afocel


81

Fédération Nationale des Entrpreneurs de Travaux Francoise Pasquier Agricoles France. Phone: +33 3 01 932955 Ruraux et Forestiers (FNETARF) Fax: +33 3 01932955 4, Rue de Alliages, E-mail: [email protected] F 254 90 Dampierre les Bois France Forestry Contracting Association (FCA) Barrie Hudson Dalfling Farm, Blairdaff Phone: +44 1467651368 AB51 5LA Inverurie Fax: +44 1467 651595 United Kingdom E-mail: [email protected] Forestry Commission Bill Jones Forest Research Agency Phone: +44 1387 860264 Technical Development Branch, (TDB) Fax: +44 1387 860386 Ae Village E-mail: [email protected] DG1 1QB Dumfries United Kingdom Norwegian Forest Research Institute (Skogforsk N) Tore Vik Høgskoleveien 12 Phone: +47 64 94 90 00 N 2432 Aas Fax: +64 94 29 80 Norge E-mail: [email protected]

[email protected] Kuratorium für Walt und Forstarbeit (KWF) Günter Weise Sprembergerstrasse 1 Phone: +49 6078 78513 D 648 23 Gross-Umstadt Fax: +49 6078 78550 Germany E-mail: [email protected] Albert-Ludwigs University Siegfried Lewark Forest Utilisation and Work Sciences (Uni-Freiburg) Phone: +49 761-2033764 Werderring 6 Fax +49 761-2033763 D 790 85 Freiburg E-mail: [email protected] Germany Qualifizierungsfonds Forstwirtschaft e. V. (QfF) Jürgen Kumm Ludwig-Erhard Strasse 8 Phone: +49 561 9354110 D 341 31 Kassel Fax: +49 561 9354141 Germany E-mail: [email protected] Warsaw Agricultural University Piotr Paschalis Department of Forest Utilisation (SGGW) Phone: +48 22 849 70 74 Rakowecka 26/30 Fax: +48 22 849 13 75 Pl 02 528 Warzawa E-mail: [email protected] Poland


82

Quality Assurance Group, QAG UN-International Labour Organisation Peter Blombäck 1211 Geneva 22 Phone: +41 22 799 7808 Switzerland Fax: +41 22 799 7967 E-mail: [email protected] National Institute of Occupational Health Bo Veiersted PO Box 8149 Dep Phone: +47 23195100 N-0033 Oslo Fax: +47 23195205 Norway E-mail: [email protected] Work Packages WP No. 1: Machine operators’ social-, safety and health situation

WP No. 2: Bench marking of Work Organisation in Logging Operations

WP No. 3: European ergonomic guidelines for forest machines

WP No. 4: European recommendations for ergo-efficient mechanised logging operations

Output from ErgoWood Main reports Implementation and the socio-economic impact of mechanisation in

Poland and France.

The machine operator’s social, safety and health situation and causal factors.

Implementation of good work organisation including control systems with cost/benefit analysis methods.

European ergonomic guidelines for forest machines, including measuring methods.

European recommendations for ergo-efficient mechanised logging operations.

An implementation plan for ErgoWood.

Design of the reports Promote the social dialogue between the parts involved in logging.

Help to find a structure in that dialogue.

Give advice about what is important.

Guide where and how to get good information.


83

Management structure

Management structureQuality Assurance Group Project manager Administrative

Scientific co-ord.

Co-ordinator B. Veiersted, STAMI, Norway P. Blombäck, ILO J Johansson

S Gellerstedt SLU/Sweden

WP4 I Johansson SLU/Sweden

D3.4 Bohlin, SLU

D8 Weise, KWF

D5 Weise, KWF

WP1 T Vik Skogforsk, Norway

WP2 E. Lidén, DELO

D1.1Vik,Skogforsk D1.3 Weise, WP3

D2.1 Winkel, D2.3 Weise, G. Weise

(SLU/Sweden)

KWF, Germany D3.1 Hudson, FCA

D4 Winkel, NIWL

D3.3 Weise,

D2.4 Bohlin, D1.4 Bohlin,

D7 Paschalis,

D6 Bigot, AFOCEL

D9 Vik, Skogforsk D13a Weise,

D10a.Vik,Skogfors D1.2 Lewark,Uni-

D11b Lidén, DELO

D3.2 Jones, TDB

D2.2 Lewark,Uni-D10b.Vik,Skogfors

D12a Jones,

D12b Bohlin, D12c Bohlin, D13b Johansson, SLU D11a Lidén, DELO

D16 Kumm, QfF

D14 Lewark,Uni-F

D15 Johansson,


84

Contracting of forest services today and in the future

O. Hultåker, F. Bohlin, and S. Gellerstedt MSc For, PhD For, and Associate Professor, Swedish University of Agricultural

Sciences, Department of Forest Products & Markets Uppsala, Sweden

Summary The aim of this pilot project was to examine the organising of logging, the business situation for forest machine entrepreneurs, and to investigate the po-tential for development. Fifteen interviews were carried out with entrepre-neurs, customers, machine manufacturers, and a forest machine operator em-ployee, based on strategic samples.

The main business of the entrepreneurs is mechanised logging. We identified 9 secondary business opportunities, which may improve income and work en-vironment. We also analysed the strategies behind these new businesses and found 5 strategies meeting needs within the company or from external sources. The integration between customers and entrepreneurs vary. Customers seem to prefer dealing with larger entrepreneurs and having full-time contracted teams. Machine manufacturers invest substantially in developing good relationships with entrepreneurs both in their machine trials and by offering improved fin-ance and maintenance services.

The entrepreneurs, professionals with local knowledge of forest owners, forest and stand characteristics, and local demand for specific wood qualities, have developed a range of new services. Given the diminishing numbers of forestry personnel in forest industry, these services may, tomorrow, come in high de-mand. Continued project research using action research methods will focus upon the prerequisites and consequences of developing these new services.

Keywords: Action research, entrepreneur, contractor, logging, strategy.

Introduction Contracting of logging services in Sweden has increased drastically since the 1970’s. Presently, contracting accounts for at least three quarters of the mecha-nised logging (Lidén 1994, Synwoldt 2001). Profitability in these often small enterprises used to be good. Nowadays it is frequently claimed to be problem-atic (Lundberg 2000). Monotonous tasks make for a poor work environment (Andersson et al. 1999).

Is there a new way of organising forestry activities including logging? This ex-ploratory qualitative research project aimed at investigating the potential for development in the logging entrepreneur business. What would the future role


85

of the entrepreneur be? What would the consequences be for work environ-ment and profitability? Might a closer co-operation with machine manufactur-ers prove mutually beneficial regarding e.g. machine development and market issues as well as machine servicing and financial services?

Materials and methods A statistically non-representative stratified sample was made to identify re-spondents from four forest subsectors – 1) successful forest machine entre-preneurs, 2) representatives of forest companies and industries, 3) forest own-ers’ associations, and 4) forest machine manufacturers. In all, fifteen interviews were carried out, nine of which targeted the entrepreneurs, four of which tar-geted forest companies and industries and forest owners’ associations, two aimed at forest machine manufacturers, and one at a forest machine operator employed by a large forest company. The interviews were carried out in 2002.

Results The main business of the entrepreneurs is mechanised logging. We have identi-fied 9 other business opportunities, which may give separate incomes and bet-ter opportunities for work rotation. One or more of the interviewed entrepre-neurs are engaged in planning of nature conservation, mechanised scarification and planting, long-term logging planning, road transports, market co-ordina-tion activities, manual logging, or developing of administrative IT-systems. By broadening their scope of activities or by playing a co-ordinating role the en-trepreneurs may obtain direct and indirect profits. However, new business op-portunities do not necessarily imply improved work rotation, since the entre-preneur employees are often specialised in operating logging machines. New tasks will often be performed by the entrepreneurs themselves, other personnel not being machine operators, or sub-contractors In order to offer employees a safer source of employment, entrepreneurs may co-operate with other logging entrepreneurs or entrepreneurs in other businesses (Hultåker 2002).

The entrepreneurs’ secondary businesses have emanated from different local opportunities and from their competence and interests. We have identified five active strategies behind these businesses (Table 1) deployed to meet needs ei-ther within the company or from external sources. The most important reason for refraining from developing new services is lack of adequate prices.

The integration between customers and entrepreneurs varies. Many cus-tomers seem to prefer having full-time contracted logging teams. The services demanded by the customers vary. In the future, they would like to give the entrepreneurs a broadened responsibility but remark that this could call for new entrepreneur competence. Customers also claim to prefer fewer but larger entrepreneurs providing both logging and other silvicultural services. However, diversifying must not infringe upon log-ging specialist competence. The customers seem to consider rationalising of logging to be a purely technical question.


86

Table 1. Identified active strategies behind entrepreneurs’ secondary businesses. Strategy Reason Provide continuous occupation for personnel or machine resources

Create alternative occupation Utilise marginal resources

Diversify the business Curiosity Improve the profitability Vary the work tasks

Administrative reasons Render own work more effective Simplify own work

Profit on the company’s competence Demand from customers Increase capacity Demand from customers Forest machine manufacturers’ technical development proceeds with little contribution from the forest companies but in co-operation with entrepreneurs on the forest site. The manufacturers aim to strengthen their ties to the ma-chine buyers and develop financial and maintenance services to this end. The service agreement may serve to facilitate interchange of competence and expe-riences between machine operators and service personnel. The manufacturers also emphasise that entrepreneurs need to develop their businesses.

Discussion We interviewed successful forest machine entrepreneurs to identify new ways of organising forestry. Therefore the solutions we have found are not neces-sarily applicable to all entrepreneurs. Continued research will address this issue. The interviews with customers and machine manufacturers were even more limited. We need in-depth knowledge about technical, economic, and organ-isational questions raised when entrepreneurs develop their businesses as well as the effects of different solutions. Action research offers methods for this.

Entrepreneurs’ diversification has been the outcome of different strategies which could benefit a larger number of firms. One future possibility of devel-opment lies in integrating logging with nearby stages in forestry work, thus making it possible to provide a range of forest services and to strengthen the entrepreneurs’ position. Closer co-operation with other entrepreneurs is an-other future possibility, which may give the same outcome. Integration or co-operation could facilitate an ambition among entrepreneurs to take a stronger responsibility for supplying industries with wood. In the future new forms of collaboration within logging business may develop as new actors enter the arena and existing actors change their positions. What internal and external prerequisites are required for developing new services within the contracting business? Under what conditions may diversified business be profitable?

The variation of work tasks for individuals, e.g. work rotation, depends not only on the existence of different tasks but also on how work is organised. The experience of profitability from work rotation influence its existence. Under what circumstances will new services within the forest contracting business improve work environment?


87

Conclusion Our research indicates that the future contracting of forest services may go well beyond logging operations. The respondents we encountered, professional forest entrepreneurs with local knowledge of forest owners, forest and stand characteristics, and local demand for specific wood qualities, showed that they have been able to invest in and develop a range of new services. Given the di-minishing numbers of forestry staff in existing organisations, these services may, tomorrow, come in high demand. We will use action research methods to continue studying under what circumstances diversified business yield impro-ved work environment and profitability as well as what internal and external prerequisites that are required for developing new services.

Acknowledgement This study has been financed by Vinnova – Swedish Agency for Innovation Systems.

References Andersson, W., Ekengren, M., Lindbäck, L., Lundmark, L., Olsson, A., Persson, G., &

Synwoldt, U. 1999. Uppföljning av skogsbrukets projekt ”Utveckling av arbets-miljö och produktion”. Arbetarskyddsstyrelsen, Solna. Rapport 1999:2. (Follow-up of Forestry’s Project “Development of Work Situation and Production”. In Swed-ish)

Hultåker, O. 2002. Forest Contracting Today and in the Future – A Qualitative Study of Logging Contractors’ Activities and their Visions of the Future. Sveriges lant-bruksuniversitet, Institutionen för skogens produkter och marknader, Uppsala. Ex-amensarbete 9. Master thesis. (In Swedish, summary in English.)

Lidén, E. 1994. Machine Inquiry -93 – Harvesting Systems, Ownership, and Volumes in Swedish Forest Enterprises; Outcome 85/86 and 92/93 and Prediction for 97/98. Sveriges lantbruksuniversitet, Institutionen för Skogsteknik, Garpenberg. Uppsatser och resultat 269. (In Swedish, summary in English.)

Lundberg, L. 2000. Financial Key Figures for Forest Contractors. Sveriges lantbruks-universitet, Institutionen för skogsekonomi, Umeå. Unpubl. (In Swedish, abstract in English.)

Synwoldt, U. 2001. The Swedish Work Environment Authority and its Initiatives Re-lating to the Work Environment in Swedish Forestry. Sveriges lantbruksuniversitet, Uppsala. Acta Universitatis Agriculturae Sueciae Silvestria 186. Diss.


88

Profile Oscar Hultåker is graduate student at Swedish University of Agricultural Sciences where he currently does research on developing new forms for forestry contracting which will permit entrepreneurs diversifying their tasks. Oscar Hultåker obtaind his MSc in Forestry in 2002. He has also studied social sciences at Uppsala University and has been employed to do qualitative market research at SKOP, Skandinavisk opinion ab.

Oscar Hultåker SLU, Inst. f. skogens produkter och marknader box 7060 SE-750 07 Uppsala Sweden +46-(0)18-67 25 44 [email protected]


89

Management Plan


90

Skogssällskapet’s business concept ”As an independent manager, Skogssällskapet will co-operate with forest property owners in order to create a good yield and develop the future values of forest properties”. Based on the business concept, our task is to undertake professional forest management that derives from the customer’s objectives and prerequisites. All categories of woodland owners are to be found among our customers. Of the total area of over 450 000 hectares of woodland that Skogssällskapet manages, just under half of the area is privately owned. Other categories of owner are local authorities, foundations, companies and the state. Management Plan in summary Skogssällskapet’s planning system for professional silviculture, called Management Plan, is a further development of the traditional forest management plan. Using current information technology, woodland owners can be part of a planning network in order to share information and to actively participate in the management decision process. Basic information from different sources is linked together and illustrated in the map. The property’s different prerequisites are analysed in terms of the owner’s objectives. Short- and long-term financial returns are weighed against the property’s other values. The woodland owner’s overall objectives generate silvicultural descriptions and forestry and multi-use programmes. The silvicultural description refers to different multi-use elements. The property’s productive stands are further analysed economically. Felling yields are calculated based on the local market. Operational planning subsequently follows in collaboration between the woodland owner, land agent and customer economist. The woodland owner can follow the operational activity using the Internet, with basic data on budget, accounts and various follow-ups. The Management Plan is a traditional forest management plan that can be extended in accordance with the requirements of the owner into a comprehensive planning tool. It provides access to constantly updated information on the owner’s woodlands using a computer connected to the Internet. The Management Plan is the ideal tool when several persons are working with the same information. The Management Plan is an unlimited planning network for forest management, in which woodland owners can select from all the services that Skogssällskapet has to offer.

References System supplier Berget Systemdesign AB, Falun Sverige SA-fmu/coc-1057 Trademark 1996

Forest Stewardship Council A.C. Economic model inspected by Lennart Eriksson, senior lecturer at the Swedish University of Agricultural Sciences


91

Planning Network Management Plan

Organisation Woodland owners are offered help with comprehensive planning for their forest property. The planning network is made up of parties who can deal with the entire chain from plan production to forest management.

The Management Plan is a live tool. The information is updated with new inventories and bulletins so that the owners are able to take care of decision making and distribution of information for all objectives pertaining to their operations on the property.

System The fact that the Management plan is server-based means that it is excellently suited for company-style management comprising several properties and with several persons involved. The plan is also suitable for individual properties. Access to the system is via the Internet either from Plan-i for information and fairly simple editing or Plan-c, the complete system on an individual’s own PC with map editing, analyses etc. The user is governed by authorisation and roles according to the below.

Plan production Information Control of measures Change register Change the map Specialist Control the system

Plan production Information

Owner

Plan refinement Management

Planning network with landowner and experts in forest management, planning and the planning system. Plan hotel with connection over the Internet for standard functions or complete installation for specialists.


Plan production and forest management Plan specification Before production of the plan decisions are made on content and level of precision with which the inventory is to be implemented. Using quality control, the products are followed up on ordering. Further choices are offered in the form of FSC-adjustments and production assessments.

Forest inventory

Field inventory In the forest inventory the woodland is divided into similar compartments. Facts about them are estimated. For more precise information a forest assessment is used. The map is digitised and the information is inspec-ted using error checking.

Assessment

Management As the basic data for the budget, the annual forest management plan is displayed. The owner participates in the decisions using the Internet.

Production is planned in the felling plan. Measures are run through supply and machine planning. Implementation is governed by compartment-directives with adjustment and map diagrams. The Management Plan produces basic data with descriptions of compartments etc.

Management plan

Felling plan

Following-up The woodland is updated using the feedback report, and the next measure is proposed. Consideration shown to natural and cultural objects is reported. In Plan-i, operations can be followed up. An economic report is produced on an ongoing basis, containing the period’s balance, timber prices etc. At the end of the year, an appendix to the annual accounts is produced with implemented measures. An annual account with key figures for the quality that has been achieved in production during the year.

92


93

Refined Management Plan Overall multi-use and landscape plan A traditional forest management plan is primarily a compartment register with proposals for measures based on current silvicultural programme. In order to obtain an overall grasp of the maintenance of a property including objectives over and above those of the forestry sector, the plan needs to be supplemented, a so-called multi-use plan. The owner’s overall objective generates silvicultural descriptions and programmes for forestry and for whichever multi-use sectors are required. The silvicultural description contains the woodland owner’s overall objectives based on financial, environmental and social perspectives. Areas of special consideration and restrictions are observed. The first stage in dividing the woodland into zones takes place on a site map, based on the future vision and the property’s prerequisites for different multi-use objectives. To adjust the zones, the map showing the current state of the woodland is used. The core area for the respective zone is maintained intact.

Multi-use layer The multi-use elements that are to be managed operationally along with proposals for measures that are to be taken, are placed in a multi-use register and on the map. The woodland is reshaped by means of controlling measures in for example a recreation zone.

Forest Maintenance A long-term simulation based on biological prerequisites is produced for the zones designated for growth of timber. A silvicultural programme for each class of objective, as well as proposed measures governs the calculations. The result is levels set for future felling and woodland condition.

Tactical felling plan A tactical felling plan is produced to prioritise the measures financially. Felling is controlled in terms of time and space on the basis of a commercial perspective. The calculations are based on personal experience and on the local market with regard to selection, using revenues and felling costs.

The complete Management Plan The multi-use section of the plan ensures the value outside the forest sector. The property is FSC-certified as support for an economic, environmental and social use of the woodland.

Forestry operations are based on strategic silviculture from a commercial perspective.

The plan is quality assured in terms of its objectives and continuously updated.


94

Profile Anders Braide Specialist with wide-ranging practical experience. Certified forester with degree (1984) specialising in administration. Line work in forest management, timber and transportation. Area manager for forestry administration service. Educator on the forestry science programme. Organisation and finances developer. Product developer in forestry planning in the field of management. Skogssällskapets Förvaltning AB (the Society of Forestry’s Management AB) since 1996. Skogssällskapets Förvaltning AB Tel +46 31 335 66 00 Stora Torp Fax +46 31 335 89 07 Box 5083, 402 22 GÖTEBORG Web www.skogssallskapet.se


95

How to protect forest fire fighters of heat stress?

Elías Apud and Felipe Meyer Unit of Ergonomics, Faculty of Biological Sciences,

University of Concepción, Chile

Heat exposure associated to high energy expenditure due to manual work is one of the main problems that affects forest fire fighters. As shown in figure 1, because the risk of direct contact with flames, they work completely covered with cloth and different protective devices.

Figure 1. Fire fighter working withcloth without ventilation

Considering the high radiant heat and air temperature to which they are exposed, the only way they have to get rid of the heat they produce during muscular work is to evaporate sweat. However, the cloth they use represents a barrier difficult to overcome.

24

52

124

1.0-1.5 1.5- 2.0 2.0-2.5 2.5-3.0

sweat/hour (kg)

Figure 2. Ranges of sweat rate offire fighters building lines (n=48)

Sweat rate of fire fighters during line construction was measured and it was found that the average ost was 1.69 liters per hour, ranging from 1 to 3 liters per hour, which is extremely high (see figure 2). 0

10

20

30

40

50

60

% o

bser

vatio

ns

l


96

An essay carried out to determined how much sweat was left in the cloth, showed that 40% of the sweat produced remained in the cloth (table 1).

Table 1. Results of one essay carried out in 8 fire fighters to detect the amount of sweat which remained in the cloth. Essay of 1 hour line construction* Body weight (kg) Body weight dressed before the essay (kg) 80.4 Body weight dressed after the essay (kg) 79.4 Liquid ingested during the essay (kg) 0.68 Liquid eliminated (kg) 1.68 % evaporated 59.6 % retained in the cloth 40.4 * Essays of four periods of 15 minutes of work and 3 minutes rest after each period of work.

Based on these findings, shirts and trousers with special openings for venti-lation, shown in figure 3, were designed. In the case of power saw operators careful considerations were made because they have to wear cutting protective devices.

Figure 3. Cloth for firefighting with ventilation inshirt and trousers

Another finding was that they drink small amounts of water, partly because it became hot after a short while and also because they only take a 1 liter capacity billican. For that reason, a special bag made of aluminum material was design-ned to carry the billican. The billican, the original bag and the newly designed bag asre shown in figure bag is shown in figure 4.


97

a b c

Figure 4. a) traditional plastic billican; b) bag used to cover the billican;newly design bag made of aluminum material.

The new bag was tested under experimental conditions and is was found that they very effective to keep the water fresh. As shown in table 2, when these water bottles were exposed to 52°C radiant heat for one hour, the temperature of the water increased only 3°C when cover with the newly designed bag.

Table 2. Results of one essay carried out to study the effect of a bag made of aluminium material to keep the water cool Common bag Aluminium bag Globe temperature at the beginning 39 39 Globe temperature at the end 52 52 Water temperature at the beginning 12 13 Water temperature after half an hour 21 15 Temperature difference after half hour 9 2 Water temperature after one hour 28 16 Temperature difference in the second half hour 7 1 Total difference in one hour 16 3 In addition, to study the effect of water drinking, 2 essays during line construc-tion were carried out for one hour time. In the first, the workers were told to drink whenever they wanted but only one out of 10 did it. Another essay was carried out with the same fire fighters, but they were asked to drink 200 cc be-fore start working and they kept drinking about the same amount every 15 minutes. As it is shown in figure 5, the results showed that when they drank water their performance in line construction improved 8.4 % and that their cardiovascular load decreased, as average, 8 percent.

2nd rs Forest Engineering Conference Poste

98

drinking water without drinking

75.5 % CL

68.4% CL

816m2/hour

753m2/hour

Figure 5. % Cardiovascular load (% CL) and output (m2/hour) as a result of an essay to test the effect on performance of drinking water at regular intervals and drinking when they felt thirsty Considering the importance of hydration, illustrated material, such as that shown in figure 6, was prepared to educate workers on how to drink and what to drink in fires where they have to stay exposed to heat for different times. Also material was developed to provide guidelines for the heads the brigades on how to protect their workers during forest fire fighting

Figure 6. Illustrated material for training.

An evaluation carried out during one complete season showed that the changes in clothing and the new bag for keeping the water cool were very well accepted by the workers. During this summer season 3 companies adopted all the chan-ges and the training material was used to educate the fire fighters on how to protect themselves of fatigue and dehydration when facing fires out of control.


99

Profile Elias Apud Simon, MSc., PhD.

1. Present Position: Professor, Director of the Unit of Ergonomics, University of Concepción, Chile.

2. Forest Research Projects last 5 years: “Appropriate technologies for the forest sector”. Financed with state and private companies funds. General Director

“Ergonomics applications to increase operational efficiency in forest fire fighting”. Financed with state and private companies funds. General Director.

Felipe Meyer Cohen

1. Present Position: Head of Technological and Scientific Services of the Unit of Ergonomics, University of Concepcion, Chile.

2. Educational Background:

Forest Engineer, University of Concepcion, Chile 1999.

Diploma on Ergonomics, University of Concepcion, Chile 2001.

Master of Science, University of Concepcion, 2003.

3. Research Projects

“Appropriate technologies for the forest sector”. Financed with state and private companies funds. Researcher.

“Ergonomics applications to increase operational efficiency in forest fire fighting”. Financed with state and private companies funds. General coordinator.

4. Publications Apud, E., Gutierrez, M., Lagos, S., Maureira,F., Meyer, F. y Espinoza, J. (1999)

“Manual de Ergonomía Forestal”. Ed.:Valverde Ltda, Concepción, Chile.

Apud, E., Meyer, F. y Maureira, F. (2002) “Ergonomía en el combate de incendios forestales”. Ed. Valverde, Concepción, Chile, 2002.

Apud, E., Gutierrez, M., Maureira, F., Lagos, S., Meyer, F. y Chiang, M.T. (2003) “Guía para la evaluación de trabajos pesados”. Ed.: Trama, Concepción, Chile.

sf-3268 forest engineering conference - framsidor till ... · lumber retailer lumber wholesaler...

Documents