340-358 3120 a participatoryenterprise architecture methodology firstly presented by spewak and hill...

Introduction

Forest resources management in a region is typicallyconducted by individual stakeholders acting mostlyindependently. Decisions are thus idiosyncratic. Theyare framed by individual economic, ecological and/orsocial objectives within specific institutional arrange-ments and organizational structures (e.g. non-industrialprivate forestland owners (NIPF), industrial owners,associations, private companies or public administra-tion). The decision-makers organizational context dic-tates the spatial-temporal scales used for forest planning.The decisions reflect the stakeholders unilateral per-ceptions of the regional forest sector, often based on

imperfect, outdated and un-shared information. So-phisticated models and tools are available to supportindividual decision-making and help address currentas well as emergent forest management planning pro-blems (e.g. Borges et al., 1999; Falcão and Borges, 2002;Falcão et al., 2006; Palahi and Pukkala, 2003; Nieu-wenhuis and Tiernan, 2005; Dias-Balteiro and Romero,2008; Constantino et al., 2008; Toth and McDill, 2008;Vainio et al., 2009; Pukkala et al., 2009; Costa et al.,2010; Forsell et al., 2009; Ferreira et al., 2011; García-Gonzalo et al., 2011).

However, stakeholders’ decisions do impact oneanother leading to complex interaction networks thatthese tools can hardly acknowledge, if used withinunilateral frameworks. For example, the selection offorest species by a NIPF will affect later in time theavailability of raw material to the neighbouring trans-

A participatory approach to design a toolbox to support forestmanagement planning at regional level

A. F. Marques1,2*, J. G. Borges2, J. Garcia-Gonzalo2, B. Lucas3 and I. Melo41 INESC Porto. Campus da FEUP. Rua Dr. Roberto Frias, 378. 4200-465 Porto, Portugal

2 Centro de Estudos Florestais. Instituto Superior de Agronomia. Technical University of Lisbon.Tapada da Ajuda. 1349-017 Lisbon, Portugal

3 EFIMED Mediterranean Regional Office of the European Forest Institute. Castella, 33. Barcelona, Spain4 ACHAR-Associação dos Agricultores da Charneca. Rua Direita de S. Pedro, 152. 2140-098 Chamusca, Portugal

AbstractAim of the study: Forest management planning in a region typically involves multiple stakeholders. Decisions

processes are idiosyncratic, driven by individual goals and supported by segmented forest-based information.Nevertheless, stakeholders’ decisions do impact one another leading to complex interaction networks wherecommunication, cooperation and negotiation play a key role. This research addresses the need to develop decisiontools to support these roles. Emphasis is on the integration of participatory planning tools and techniques in thearchitecture of a regional decision support toolbox.

Area of the study: The proposed approach was applied in the Chamusca County in Central Portugal although it iseasily extended to other regions.

Material and methods: This research proposes an Enterprise Architecture methodological approach to design atoolbox that may address distinct stakeholders’ interests and decision processes, while enabling communication,cooperation, negotiation and information sharing among all those involved in the regional interactions network.

Main results: the proposed approach was tested in a regional network involving decision processes and informationshared by 22 entities clustered into 13 stakeholders groups, including industrial owners, and non-industrial privateforestland owners (NIPF) —acting individually or grouped into associations and federations—, national and regionaloffices of the forest authority, forest services providers, non-governmental organizations and research centers. Resultssuggest that the proposed approach may provide a toolbox that may effectively address stakeholders’ decision processesand goals and support the regional interaction network.

Key words: forest management; multiple stakeholders; decision support systems; enterprise architecture; partici-patory process.

* Corresponding author: [email protected]: 18-05-12. Accepted: 29-04-13.

Instituto Nacional de Investigación y Tecnología Agraria y Alimentaria (INIA) Forest Systems 2013 22(2), 340-358 Available online at www.inia.es/forestsystems ISSN: 2171-5068http://dx.doi.org/10.5424/fs/2013222-03120 eISSN: 2171-9845

formation centres, further impacting in the goals rela-ted with the forest species coverage set in regional fo-rest plans by the public administration. Similarly, NIPFharvesting decisions are impacted by demand require-ments and will in turn affect the regional availabilityof harvest equipment, regional employment and land-scape configuration. Conversely, the lack of equipmentor unfavourable prices may influence the postponementof his harvesting decisions. Moreover, the social andecological impacts of these decisions may be conside-red unacceptable to local communities.

The lack of acknowledgement of these interdepen-dences within the management planning frameworkmay lead to unnecessary conflicts and sub-optimalresources utilization (Grimble and Wellard, 1997; Mar-tins and Borges, 2007).

The development of regional toolboxes that may in-tegrate models/methods, procedures/documents andtools currently available to support individual decision-making as well as facilitate communication, coope-ration, negotiation and information sharing techniquesbetween stakeholders thus emerges as a pertinent re-search problem. To our knowledge, no such regionaltoolboxes (RgTbx) have been developed. Our RgTbxconcept has similarities with group decision supportsystems (GDSS). The latter, rely on collaborative tech-nologies and enable distributed meetings and groupwork. Korpela et al. (2001) applied a GDSS-basedapproach to the analysis of the strategy of forest indus-tries. The RgTbx concept extends these functionalities.

The number of agents involved in regional forestplanning suggests the need to use participatory planningtechniques to strengthen stakeholders’ involvement inthe toolbox design. This will be influential for adequaterepresentation of communication and cooperationprocesses to be supported by the toolbox. In fact, par-ticipatory planning techniques have been successfullyin the context of collective forest planning, particularlywhen the objectives and planning processes are ill-structured or the information is scarce (e.g. Martinsand Borges, 2007; Ananda and Herath, 2003; Kurtillaand Pukkala, 2003; Purnomo et al., 2005; Kangas etal., 2008; Hjorts, 2004). These techniques may also beused within a multi-criteria decision analysis frame-work to help assess strategies and plans (e.g. Díaz-Bal-teiro and Romero, 2008; De Steiguer et al., 2003; Sheppardand Meitner, 2005; Tecle et al., 1998; Schmoldt et al.,2001; Nordström et al., 2009).

Finally, and of specific interest to this research, par-ticipatory planning techniques may be very helpful for

information systems design, e.g. within an EnterpriseArchitecture approach. Ribeiro et al., (2005) and Mar-ques et al. (2010a) applied such techniques with thetechnical profiles of Portuguese pulp and paper compa-nies during the workshops for designing corporativeforest management decision support systems. Marqueset al. (2010b) and Marques et al. (2012) also used par-ticipatory planning processes to design an interopera-bility platform between the multiple information sys-tems that are used by those involved in the wood supplychains.

In this research we propose a novel stakeholders en-gagement plan that is a participatory planning processused in the context of an Enterprise Architecture fra-mework, to involve the stakeholders in the design aregional toolbox (RgTbx). The stakeholders’ engage-ment plan aims to explicit and document their know-ledge, concerns and requirements. No a priori assump-tions are made about individual decision processes andnetwork interactions so that the toolbox may effectivelyaddress stakeholders’ needs and successfully accommo-date the human dimension into its development.

The proposed approach is an extension of theEnterprise Architecture methodology firstly presentedby Spewak and Hill (1992) and applied in forestry con-texts by Ribeiro et al. (2005) and Marques et al.(2010a). Specifically, the approach relies on modelinginformation, forest decision processes and interactionsnetworks in the framework of Process Architectureworkshops with the stakeholders. This approach thusprovides a collective and consensual vision represen-ting both the individual decision processes and the in-teractions networks. Morevoer, it enables the identi-fication of the RgTbx components, including the toolsrequired to support individual forest planning pro-cesses and the data elements handled by the stakehol-ders. A complete specification of the RgTbx is influen-tial for the alignment between the forest managementdecision processes and the IT function (Sousa and Pe-reira, 2005; Sousa et al., 2005).

The proposed EA methodology was tested on Cha-musca County, located in Central Portugal. The stake-holders involved in the design of the RgTbx includedindustrial owners and NIPF acting individually orgrouped into associations and federations. It furtherincluded national and regional off ices of the forestauthority, responsible for managing public forests aswell as providing regulatory frameworks for forestmanagement planning. Representatives of forest servi-ces providers, forest industries, forest investment funds,

Participatory design of RgTbx 341

non-governmental organizations and research centerswere also involved.

Material and methods

The Chamusca county case study

The Chamusca county is a rural and low populationdensity municipality, extending over 74,599 ha in theCentral Portugal. Forests extend over 51% of the coun-ty territory. Eucalypt and maritime pine plantationsextend over 62% of the county forest area while corkand holm oak multi-functional forests occupy 35% ofthis area. The remaining 3% corresponds to protectionareas.

The forest is this region is predominantly private.Pulp and paper companies and a few large-scale NIPFmanage 73% of the forestland while the remaining areais held by more than 2,200 NIPF, some with holdingswith less than 1 ha. These stakeholders can act indivi-dually or grouped into forest associations and federa-tions. Typically, NIPF sell stumpage and cork to localtrade entrepreneurs. Often, the latter own the harves-ting equipment and rely on local workers hired for theharvest season. The transportation of forest productsis typically outsourced to logistic operators or indi-vidual carriers.

Forest operations are regulated by the regional officeof the forest authority according to the regional planand forestry policies. The regional office is also res-ponsible for managing the public forests in the county.Recently and as a response to the 2003 wildfires thatburned 20 × 103 ha of the county’s territory, the localmunicipality also plays a key role in developing andsupervising forest wildfires prevention plans as wellas in coordinating the forest wildf ires suppression

efforts. Other stakeholders include forest investmentfunds, non-governmental organizations, local commu-nities and forest research agencies.

Methods

The stakeholders’ engagement plan, anchored in theEnterprise Architecture methodology, consisted in fourmain stages for the design of the toolbox for regionalforest planning (Fig. 1). The first stage of this partici-patory process tackled the selection of stakeholders.The research team responsible for the design of theRgTbx identified a key stakeholder in the region. Thelocal forestland owners association ACHAR, acted asa local project promoter and helped in the selection ofthe relevant entities i.e. forest stakeholders in thecounty.

These entities were clustered into groups and func-tional categories to be addressed in separate processarchitecture workshops. Afterwards, at least one repre-sentative designated by each entity took part in thekick-off meeting. One forest practitioner from ACHARwas further included in the project team in order tostrengthen the contacts with the stakeholders andprovide technical support to the moderator during theworkshops.

The second stage of the proposed engagement planencompassed a series of Process Architecture work-shops to identify individual decision processes. Theseworkshops used the post-It method for designing a 3-level hierarchical top-down process architecture, overtwo consecutive half-day sessions. During the f irstsession, participants in each stakeholders’ group wereasked to identify, in Post-It notes, the entities they in-teracted with and to display them in the organization-centric context diagram, closer or farther from the

342 A. F. Marques et al. / Forest Systems (2013) 22(2), 340-358

Figure 1. Forest stakeholders’ engagement plan for designing the regional forest management toolbox.

center according to their business relevance (Marqueset al., 2010a).

The information flows among entities were furtheridentified in this first level of processes’ representa-tion, differentiating paper and electronic carriers. Theidentification of the information flows was influentialfor highlighting the interactions’ network among thestakeholders involved in forest management in theregion. It further provided the basis for the developingthe Business Model for each stakeholders group thatwas the second level of the processes’ representation.The Business model emphasized the business pro-cesses conducted by each stakeholders group. Only theprocesses directly related to forest management acti-vities were further detailed.

The third level processes’ representation consistedin flowcharts built in a Business Process Modeling No-tation during the second session with each stakeholdergroup. The flowcharts depicted the sequence of activi-ties performed by the stakeholder as well as the infor-mation used and produced in the course of the forestmanagement activities. The outcome of the individualPA workshops was displayed on a HTML data reposito-ry, enabling dynamic navigation throughout the repre-sentations, providing easy access to the objects des-criptions and semi-automatic reporting.

The stakeholders were asked to validate the out-comes of the PA workshops. At the time of validation,stakeholders were further asked to reply to an open-end questionnaire on current and future perspectivesabout their role within the regional forest managementplanning framework. The questionnaire addressed 21items in five main groups e.g. the forest managementplanning setting, the resources allocation, the imple-mentation of forest operations, the usage of compu-terized-tools, and other context elements. The threef irst groups encompassed management planningactivities. The fourth addresses the way these activitieswere automated and supported by computerized toolswhile the f ifth included specif ic regional networkelements. The answers to the questionnaires providedvaluable information to characterize the managementplanning problems that were prevalent in the region.Furthermore, the questionnaires highlighted the rolesand the activities conducted by each stakeholder groupto tackle them.

The third phase of the engagement plan used the pre-vious results for the integrated stakeholders’ analysis.For this purpose, the research team merged individualcontext diagrams into a complex integrated context

diagram. This was influential to develop the RegionalForest Management Planning (RFMP) Framework aswell as to identify the roles played in it by each stake-holder at Chamusca region. The four-level integrateddiagram displayed the main decision-makers at thecenter. The stakeholders involved in consultation, re-gulation or implementation of the decision weredisplayed at the next level. The stakeholders that arejust informed about the decisions were representednext. All other entities were represented at the levelfarther from the center. The characterization of rolesplayed by each stakeholder in the RFMP Frameworkwas based on the generic responsibility assignmentmatrix (e.g. PMI 2011), as well as on the answers toindividual questionnaires.

The results this third stage helped mapping indi-vidual decision processes in the overall framework offorest management planning in the Chamusca county.They further provided information to develop thecommunication and cooperation mechanisms neededto support the regional interactions networks, thuscontributing to enhance data sharing among the entitiesengaged in forest planning in the region.

The fourth stage identified the RgTbx components.During this phase, the research team listed the data anddecision support tools used or required by the stake-holders groups. Additionally, the team developed a newdata ownership matrix to show how each stakeholdergroup handled each data and information element (e.g.create, read, update and delete). This matrix was drivenfrom the individual decision processes and helpedidentify the actual data and information needs by eachstakeholder group. It further suggested techniques forenhancing communication, cooperation, informationsharing and exchange within an adequately regulatedaccess to the RgTbx. The results were subjected to va-lidation by the representatives of the stakeholders groupsand further discussed in the project final meeting.

Results

The first stage of the stakeholders engagement planled to the selection of 22 entities that were classifiedinto 13 stakeholders groups and 4 functional categories(i.e. private sector, public sector, non-governmentorganization, research agencies). They representedabout 900 people with direct interests in the Chamuscacounty forest sector. The entities included both activeand passive stakeholders, i.e. those who affect/deter-


mine a decision or action and those that are affectedby it (Grimble and Wellard 1997, Martins and Borges2007) (Table 1). Most representatives took part on thekick-off meeting. Nevertheless, smaller-scale ownersdid not participate in this meeting. They usually do notrely on technical forestry support thus their contactswere not available at ACHAR at that moment. Yet theydid participate in subsequent PA workshops.

The private sector category included industrialowners, NIPF, their associations and federations, fo-rest-based industries, forest service providers and theforest investment fund. The public sector category en-compassed the Chamusca municipality, one parishrepresenting local communities and the forest authority(both national and regional offices). Finally, one non-governmental organization and one forest researchcenter were also selected.

The forest holding structure as well as the manage-ment objectives and practices motivated the definitionof 4 sub-groups of NIPF. Accordingly, the large-scaleintensive sub-group included owners with forestry astheir main economic activity (e.g. timber and/or corkand pine nut production) and, typically, with holdingswith more than 500 ha. The owners in the medium-

scale intensive sub-group considered forestry as theirsecondary economic activity and their holdings had anarea between 100 and 500 ha. Nevertheless they con-ducted regularly management operations according totheir forest management plan (PGF). The owners ofthe small-scale sub-group (< 100 ha) considered fores-try as a residual activity and usually did not conductany forest operations. Lastly, the multifunctional sub-group included owners that focused mostly on othernon-wood products and market services, such as mush-rooms production, apiculture, hunting and tourism.Each sub group was further characterized accordingto planning constraints and regulations (e.g. wildfireprotection and wildlife conservation frameworks) tothe management of their holding, landowner age,academic degree and technical competences.

The identification of individual decision processeswithin prevalent management planning problems,during the second stage of the engagement plan, under-lined the contribution of each stakeholder group to theoverall forest management planning framework in theChamusca County. Forestland owners play a key rolein forest management planning in the region. The PAworkshops confirmed that the sophistication of the


Table 1. Stakeholders categories and groups

Category Group Entity Nº representatives Workshops

Private Sector Non-industrial private Large-scale intensive 5 WS1forestland owner (NIPF) Medium scale 2 WS2

Small-scale 0Multifunctional 1 WS2

Forest associations (FA) ACHAR-Associação dos Agricultores 3 WS3da Charneca

Forest federations (FF) Forúm Florestal 1 WS4UNAC-União da Floresta Mediterranica 1 WS4

Industrial private forestland Silvicaima 2 WS5owner (IPF)

Forest-based industry (I) Grupo Portucel-Soporcel 1 WS6

Forest Services Provider (FSP) BETA 1 WS7

Forest Investment IFund (FIF) Floresta Atlântica 2 WS8

Public Sector Chamusca Municipality (M) Forest technical office 1 WS9Civil Protection 1 WS9

Forest Authority DRFLVT-Regional Office (FAn) 1 WS10AFN-National Office (FAr) 4 WS11

Local communities (LC) Junta de Freguesia da Chamusca 3 WS12

Non-Government Non-Governmental WWF Mediterranean Programme 1 WS13organization Organization (NGO) Office

Research Agencies Forest Research Centre (FRC) CEF-Centro de Estudos Florestais 2 WS14

Total 13 22 32 14

planning process tends to increase with the holdingsize. In fact, both the industrial owners and the large-scale NIPF usually develop strategic (long term) plansat forest level. Typically, these plans are developed bya single decision-maker, target multiple objectives andinclude both wood, non-wood productions and services.

The decision-maker managing large holdings ismost often supported by skilled forest practitionersthat configure the plan to meet the requirements of thelegally binding forest management plan (PGF). Thefirst years of the strategic plan provide the input forthe operational plan aiming to provide actual resourcesallocation, budgeting and detailed specifications forthe forest operations implementation. The operationsare further clustered into work-orders, implementedwith outsourced resources and controlled by the owner.The management planning processes are frequentlysupported by computerized-tools with adequate growth& yield models powered by updated forest inventorydata. Optimization techniques embedded on decisionsupport systems to help search for the most profitableplan have also start been used, mostly by the industrialowners (e.g. Borges et al., 1999).

The medium-scale NIPF, the managers of publicforests and the other forestland owners technicallysupported by Forest Associations address similarstrategic problems. The forest public administration isfurther responsible for the development of regionalforest plans (PROFs). These plans are the outcome ofa participatory planning process where the stakehol-ders are involved in the discussion of the main forestgoals in the region. The goals are tied to the territorythrough the definition of sub-regions that are homoge-neous according to dominant forest functions (e.g.production, conservation). The PROFs further suggestmanagement goals, species selection and prescriptionsto be adopted by individual management plans (PGF).

Small-scale NIPF usually conduct long-term planningat stand level and target wood or cork production. Bothsmall-scale NIPF and the managers of public forestsconfirmed concerns with other non-wood products andservices with an increasing importance on the region,such as fruits (e.g. pine nuts), mushrooms, fishing andhunting. Yet multiple productions are often not addressedfully by the planning process as few production func-tions are available. Decisions regarding the supply ofthese products and services are unstructured andsupported mostly by empirical insights and traditionalsilviculture models. No computerized-tools other thangeographical systems, databases or spreadsheets are

used to support those activities. Tactical and operatio-nal planning is often absent as the implementation of fo-rest operations is usually outsourced to service providers.

The role of the recently created forest investmentfund was further highlighted. The fund buys or rentsforest properties, acting like both the owner and themanager of forest properties just like in the case of theindustry. The problems faced by the fund are mainlyrelated to the acquisition of new properties. Afteracquisition the fund properties face managementplanning problems that are thus similar to the large-scale NIPF owners and the industry problems.

The identification of individual decision processesfurther underlined the role of the local forest associa-tion, ACHAR. It provides consultancy and technicalsupport to help NIPF develop their plans. Moreover, itoften provides the resources needed to implement theindividual plans. ACHAR may further impact theregional forest management planning framework as itrepresents the landowners’ interests on national andregional forestry forums.

These actors —industry, NIPF owners, forest invest-ment fund, forest association and managers of publicforests— develop individual decision processes toaddress forest management problems. Yet these pro-cesses are impacted by a large number of stakeholderswithin a vast interaction network.

This research developed context diagrams, businessmodels as well as a detailed characterization of indi-vidual processes for all stakeholder groups. Yet forconciseness we will illustrate results focusing mostlyon ACHAR. Its context diagram (Fig. 2) displayed theregular contacts of the Forest Association with 13 otherentities. For example, the regional office of the forestauthority is responsible for checking whether theindividual management plans (PGF) developed byACHAR for his associates meet the regional forest plan(PROF) guidelines. Additionally, ACHAR interactsfrequently with the Chamusca municipality in thecontext of the wildf ire prevention and suppressioninitiatives. Moreover, both ACHAR and the forestlandowners have regular contacts with forest serviceproviders as they often outsource harvesting and corkand timber transportation operations. These small-scale enterprises play a key role in the regional forestlogistics and operational planning. Nevertheless, theyseldom use computer-based tools to address their ma-nagement planning problems. The information flowsbetween ACHAR and other stakeholders are supportedmostly by oral communications and paper requests


rather than by the use of templates or numbered docu-ments on an electronic format.

The PA workshops underlined that the participationof stakeholders such as ACHAR in the RFMP Frame-work extends beyond the support to forestland ownersdecision-making. In the case of ACHAR, only two bu-siness processes in its business model (Fig. 3) weredirectly related with providing management planningservices to the forest owners (P1) or with providingsupport to their forest products commercialization(P4). The f irst process included the development offorest management plans (P1.1), forest resourcesinventory (P1.2), cartography (P1.3), forest operationsfollow-up (P1.4), forest products evaluation (P1.5),sanitary control (P1.6) and forest investment projectsformatted for application of public subsidies to forestactivities (P1.7). Other business processes by ACHARincluded the support to the regional wildfire preventioninfra-structure (P6) and non-commercial activities,such as outreach activities (P2), associates’ represen-tation activities (P3), research and development acti-

vities (P5), cooperation for local development (P7) andadministrative management (P8).

Lastly, the analysis underlined the sequence of ac-tivities usually conducted in the framework of eachprocess. For example, the development of the forestmanagement plan, according to existing forestry regu-lations, encompasses a number of activities carried outby the ACHAR forest technicians at the request of anassociate (P1.1.) (Fig. 4). According to the sub-processflowchart (third level representation) it starts with anexpedite site characterization based on data publicallyavailable (e.g. at the forest authority offices) such asforest cover maps and PROF guidelines and recommen-dations for the zone where the property or set of pro-perties is located. If needed, it may include visits tothe forest site to collect forest inventory data.

Afterwards, the property is classif ied into homo-genous management units. The identification of mana-gement goals generally takes place in a meeting withthe owner. Typically they address revenue concerns andthey thus focus on the supply of the most important


Figure 2. ACHAR context diagram.

market products (timber and cork). Yet, the ACHARtechnicians may highlight the economic benefits ofnew prescription models, alternative species and othernon-wood productions and services. Environmentaland social concerns are typically met through the com-pliance with PROF guidelines. In the end, the manage-ment plan (PGF) suggests one prescription for eachstand and provides a rough estimate of the opera-tions/investment costs.

The integrated stakeholders’ analysis conducted un-der the third stage of the engagement plan provided anoverall interpretation of the contribution of all thestakeholders in the RFMP Framework. In particular,the integrated context diagram represented 42 distinctentities and over 85 information flows exchangedamong them (Fig. 5). The complexity of this diagrammade it hardly readable outside the dynamic HTMLrepository. It reinforced the key role of the industry


Figure 3. ACHAR business model.

Figure 4. Process of preparing the Forest management plan (P1.1) by ACHAR after forestland owner request.

and the NIPF as the main decision-makers, representedat the centre of the diagram. Other decision-makersincluded the forest investment fund and the managersof public forests.

The simplified version of the overall RFMP Frame-work (Fig. 6) highlighted the main individual decisionprocesses addressed by the key decision-makers (in-dustry, NIPF, forest investment fund and regionaloffice of the forest authority) as well as their interac-tion network. There are few explicit interactions bet-ween the first-level stakeholders. They are more proneto happen in the interface between NIPF and the forestfund in the case of forest properties included in thefund. Yet, implicit interactions were reported amongNIPF, namely regarding the setting of prices in forestproducts selling agreements.

At some extent, all f irst-level stakeholders areinfluenced by the stakeholders at the second level ofthe RFMP Framework. Specifically, the industry ow-ners relies on wood supply levels agreed with the fo-rest-based industries, while the harvesting decisionsof the NIPF are conditioned by the negotiations withcork and timber trade entrepreneurs and forest serviceproviders. The wood harvested impacts the demandrequirements of the neighboring transformation cen-ters. Additionally, both forest owners and the forestinvestment fund interact with the Chamusca munici-pality for licensing their forest operations accordingto regional and national regulations (e.g. deriving fromPROF). The forest authority supports the municipality

in the application of the forest regulations and usuallyalso intervenes on some PGFs approval. This is, forexample, the case when the property is part of a collec-tively managed area (such as the ZIF) or when the ma-nager is applying for public funding.

These network of interactions mainly relies as verbalcontacts and oral agreements although document ex-change may also occur. Electronic information exchan-ges are rare, while data sharing and exchange throughintegration of information systems owned by distinctstakeholders are inexistent.

The third level of the RFMP Framework includespassive stakeholders (e.g. local communities, forestresearch centers, non-governmental organizations,forest federations, industry associations and huntingassociations). These entities do not interact so frequen-tly with the first-level stakeholders and yet they pro-vide information and support to enhance forest mana-gement planning at the Chamusca County. For example,the forest owners’ federations and the forest researchcenters conduct regularly outreach activities (e.g.training courses, experimental development and de-monstration of novel computerized-tools to supportforest planning) to address first-level stakeholders in-terests and requests.

The stakeholders groups involved on the PA work-shops further reported 20 other national and interna-tional entities (e.g. police department and internationalsectorial associations) indirectly engaged in forestmanagement planning in Chamusca. These are repre-


Figure 5. Integrated context diagram for the stakeholders involved in forest planning at the Chamusca county.

sented at the fourth level of the integrated contextdiagram.

The integrated stakeholder analysis further focusedon the roles of the stakeholders within the RFMPFramework (Table 2). Results highlighted that eachactivity involved an average of 7.4 stakeholders. As anexample, the def inition of management planningconstraints involved up to 10 stakeholders groups. Thisset included 4 f irst-level stakeholders groups (deci-sion-makers), 2 second-level stakeholders groups (res-ponsible for the setting up of constraints withinnational and regional planning) and 2 additional groupsto be consulted about this issue (e.g. forest associa-tions). At another extreme were the activities involvingonly one stakeholder group e.g. the selection of serviceproviders that are exclusively undertaken by the forestowners.

Moreover, the results reported in the role matrixconfirmed the importance of first-level stakeholders,which are directly involved in 90-95% of the RFMP

Framework problems and activities. The Chamuscamunicipality and the forest authority are the onlystakeholders that address the RFMP Framework acti-vities through an approval role. The forest researchcentre and non-governmental organizations addressthese problems and activities through a consultancyrole while the forest association may play several roles(e.g. decision-maker, implementation and consultancy).

The stakeholder groups’ replies to the questionnairehandled during the last stage of the engagement planled to the listing of 20 tools needed to support theircurrent and future individual decision processes withinthe forests management planning problems prevalentin the region (first part of Table 3). This was influentialto def ine the RgTbx components. Only 21% of thetools listed are already in use. Specifically, the growthand yield models (Q2) and the harvest planning models(Q6) are often embedded in the information systemsused by industrial owners, large-scale NIPF and forestassociations. The prescription models (q1) and forest


Figure 6. Simplified version of the Regional Forest Management Planning Framework, including the main individual decision pro-cesses of the stakeholders groups engaged in forest management planning in the Chamusca county.


Table 2. Matrix identifying the role of each stakeholder group in the regional forest management planning framework. Itlists the roles using the categories: Responsible for developing the activity (R), Approves/regulates activities (A), Imple-ments activities (E), Informed about activities (I) and Consulted/provides guidelines for activities (C). The stakeholdersgroups were: Non-industrial private forestland owner (NIPF), industrial owner (IPF), Forest Investment Fund (FIF), ForestAuthority regional office (FAr), Forest Service Provider (FSP), Forest Association (FA), Forest-based industry (I), Munici-pality (M), Forest Authority national office (FAn), Forest Research Center (FRC), Forest Federation (FF), Local community(LC) and Non-Governmental Organization (NGO)

Forest management Stakeholders groupsTotal

planning framework NIPF IPF FIF FAr FSP FA I M FAn FRC FF LC NGO

Forest management settingA1. Defining Management

Units R R R R E C C 7A2. Defining Production

Objectives R R R R C C C C C 9A3. Defining Other

Management Objectives R R R R C C C C 8A4. Defining Management

Constraints R R R R E A A C C C 10A5. Defining Conservation

Actions R R R R C C C 7A6. Defining Erosion

prevention actions, forest diseases, forest fire prevention R R R R E E A C C 9

A7. Defining Infra-structure maintenance R R R R E A C C 8

Resources allocationB1. Forest species selection E/R R R R R/E A A C C 9B2. Forest prescriptions

selection E/R R R R R/E R C C 8B3. Equipment and

operations protocol selection E/R R R R R/E 0 C C C 8

B4. Service providers selection R R R C 4

B5. Forest products usage and commercialization R R R R 0 C C C C C 10

Forest operations implementationC1. Timing of maintenance

operations R R R R 0 I C 6C2. Timing of harvesting

operations R R R R 0 R I C 7C3. Timing of cork-

stripping operations R R R R C I C 7C4. Timing of forest

inventory activities I R R R R/E R C 7

Usage of computerized toolsD1. FM optimization E/R R R R R/E C 6D2. FM tools E/R R R R R/E C R 7D3. Risks and climate

change simulator R R R R C C C 7

Context elementsE1. Local communities

interaction R R R R C 0 I C R 8E2. Forest policies and

strategies def. C C R/E C C 5

Total 20 21 20 19 3 15 2 7 15 21 11 1 2% 95% 100% 95% 90% 14% 71% 10% 33% 71% 100% 52% 5% 10%


Table 3. Components of the forest management planning RgTbx. It includes the models/methods and the procedures/do-cuments needed to support both individual processes developed by stakeholders groups and the stakeholders’ network inter-action. It further describes the way each stakeholder group interacts with each data and information elements classified asCreate (C), Read (R), Update (U); Delete (D). (*: GIS-based information); The stakeholders groups were: Non-industrialprivate forestland owner (NIPF), industrial owner (IPF), Forest Investment Fund (FIF), Forest Authority regional and na-tional offices (FAr, FAn), Forest Service Provider (FSP), Forest Association (FA), Forest-based industry (I), Municipality(M), Forest Research Center (FRC), Forest Federation (FF), Local community (LC) and Non-Governmental Organization(NGO)

Decision Support toolsStakeholders groups

TotalNIPF IPF FIF FAr FSP FA I M FAn FRC FF LC NGO

Models/methods

Q1. Forest productivity zoning × × × × × × × 6Q2. Regional growth and yield models × × × × × × 6Q3. Fruit production estimation model × × × × × × × 7Q4. Cork quality & quantity prediction × × × × × × 6

modelsQ5. Harvesting/stripping opt. Models × × × × × × 6Q6. Impacts of fertilization into × × × × × × 6

productionQ7. Forest market evolution models × × × × × × × × × × 9Q8. Product distrib. Routing, storing,… × × × × × × 5Q9. Optimal equipment allocation models × × × × × × 5Q10. Risk prediction models × × × × 4

Procedures/documents

q1. Forest operations and prescriptions × × × × × × × 7for market goods, Prescription models/ productivity classes

q2. Forest operations and prescriptions × × × × × × × 7for non-market goods and services

q3. Forest Management standard × × × × × × × × × 10procedures

q4. Conservation prescriptions × × × × × × × × 8q5. Participatory techniques for public × × × × × × × × 8

forests management and ZIF management

Other tools

t1. Training courses on planning tools × × × × × × × × × × 10t2. Training and support on non-wood × × × × × 5

products and services management t3. Collective forest equip. owning/ × × × × × 5

rentingt4. Portable devices for forest surveys × × × × × × × × × × 10t5. Online Forum × × × × × × × × × × 10

Data and information elements

i1. Municipallity Management Plan* R R R R R CRUD R 8i2. Protected Areas*, meteo.*, land uses* R R R R R R R R 8i3. Hunting areas* R R R R R CRUD R 7i4. Forest Intervention Zones (ZIF)* R R R R CRUD R RU R 8i5. Historic record of the area burned R R R R R R CRUD R 8

annuallyi6. Regional Forest Management Plan* R R R R R R CRUD R 8i7. BD Regional forest inventory data* R CRU CRU CRU CRUD CRU R 7

management standard procedures (q3) are also usedby most first-level stakeholders.

Most stakeholders pointed out to the lack of updateddata and information about forest product prices. Infact, the forest authority out-dated web informationsite on reference prices was considered very useful.

The forest association most pressing requirementwas the availability of adequate forest productivityzoning maps and regional forest production models.According to ACHAR, this would help save forestinventory costs thus reducing the forest managementplans production costs. It would further provide in-formation needed to project forest products supply overthe planning horizon.

Logistics optimization models were not directlyrequested by f irst-level stakeholders as they focusmostly on strategic forest-level planning and outsourceforest operations to second-level stakeholders. Yetaccording to the forest research centre, the use of thesemodels might contribute to reduce operations costs andincrease revenues especially in the case of integrated

supply chain management. The forest service providerscould also benefit from these models but their small-scale operation may preclude the investment in thedevelopment of these tools.

Additionally, new procedures and manuals were re-quired by most stakeholders groups in order to addressissues like biodiversity management and conservation,forest operations best-practices, procedures for productcertification and management models for conservationareas. These tools should address the specificity of themanagement planning problems prevalent in theregion. Proper training and the availability of technicalbibliography were also a major requirement.

Still in the fourth stage of the engagement plan, thecore information needed by individual processes andfurther exchanged within the regional interactionsnetwork was classified into 25 data and informationelements to be address by the RgTbx (second part ofTable 3). 48% of the data elements were already beingused, namely by first-level stakeholders. These inclu-ded mostly thematic Geographical Information System


Table 3 (cont.). Components of the forest management planning RgTbx. It includes the models/methods and the procedu-res/documents needed to support both individual processes developed by stakeholders groups and the stakeholders’ networkinteraction. It further describes the way each stakeholder group interacts with each data and information elements classi-fied as Create (C), Read (R), Update (U); Delete (D). (*: GIS-based information); The stakeholders groups were: Non-in-dustrial private forestland owner (NIPF), industrial owner (IPF), Forest Investment Fund (FIF), Forest Authority regionaland national offices (FAr, FAn), Forest Service Provider (FSP), Forest Association (FA), Forest-based industry (I), Munici-pality (M), Forest Research Center (FRC), Forest Federation (FF), Local community (LC) and Non-Governmental Organi-zation (NGO)

Decision Support toolsStakeholders groups

TotalNIPF IPF FIF FAr FSP FA I M FAn FRC FF LC NGO

i8. BD product prices RU R R R R CRUD R R R R 10i9. BD forest operations costs RU R RU R RU CRUD R R R 9i10. General info. on forest product R R R RU R R R CRUD R CRU R 11

marketsi11. BD Service providers characterization RU RU R CRUD R R R 7i12. Technical forest bibliography RU R R CRUD RU 5i13. BD Properties and forest operations* CRUD CRUD CRUD CRUD R CRUD R RU R R 10i14. BD forest investment support R CRUD R CRUD CRUD R R 7i15. Legislation analysis R R R R R R CRUD R CRUD R 10i16. Forest sectorial statistics R R CRUD R R 5i17. BD ownership structure R R R CRUD RU R R R R 9i18. BD conservation interests RU CRUD CRUD CRUD CRUD RU RU R R R 10i19. Watercourses and water repositories* RU RU RU RU CRUD R R R 8i20. Forest fires prevention plans* R R R R R CRUD R R R 9i21. Forest roads* R RU RU RU CRUD R R R R 9i22. BD Administrative info. CRUD R R R CRUD CRUD CRUD 7i23. BD equipments R R R CRUD R CRUD R R R 9i24. BD licensing requests & Infractions R R R R CRUD CRUD CRUD R 8i25. Wood demand estimates R R R R R CRUD 6

(GIS)-files (e.g. the geographical limits of main landuses at the county level (i1), the protected areas, meteo-rology data, and forest cover type areas (i2), huntingareas (i3), forest intervention zones (i4), burned areas(i5), PROF sub-regions (i6), hydrology (i19), forestwildfire prevention plans (i20) and forest roads (i21)).

ACHAR, the forest investment fund and the industryowners also used tailored information systems tomanage forest inventory data (i7), forest properties andrecord forest operations implementation (i13). Thehandling of data and information elements by stake-holders (Table 3) further highlighted the relativeimportance of stakeholders’ groups in the RFMP Fra-mework. As expected, the first-level stakeholders handledmost data and information elements (e.g. ACHARaccessed 23 data and information elements, 14 of themwith editing permissions). The forest research centercan access —on a read-only basis— the elements rela-ted with its research interests while local communitiesand non-governmental organizations cannot edit anyof the elements to be supported by the RgTbx.

The current data and information handling schemeby the stakeholders’ network may lead to redundancyand inconsistency. In fact, each data and informationelement is handled by an average of 7.8 stakeholdergroups; 75% of the elements may be created and up-dated by more than 1 group. Each stakeholder followsindependent procedures for acquiring data and produ-cing information and the outcome is not shared. Forexample, the same forest property boundary (i13) maybe delimited by a wide range of entities for distinctpurposes, e.g. by individual forestland owners tosupport forest planning, by the forest investment fundto support property acquisition negotiations and by theregional office of the forest authority for taxing pur-poses. This contributes to controversies over the actualboundaries and the property official area.

Another example is the forest inventory data (i7).Forest associations conduct forest inventory to supportthe development of PGF, while the forest researchcentre collect inventory data to support experimentalresearch and demonstration projects (e.g. developmentor adjustment of growth and yield models). Additio-nally, the forest authority conducts a periodic nationalforest inventory evaluation. The inventory protocolsare often different and the results are not promptlyshared by stakeholders.

The data and information redundancy problem isfurther compounded by the way each stakeholderaccesses the information managed by another. Access

privileges and procedures vary and this leads to signi-f icant differences in the quality and quantity of in-formation used to conduct activities in the individualprocesses. As a consequence the outcomes of similarmanagement planning processes may be different thuscomplicating in turn the comparison between distinctplanning exercises.

Therefore, communication, cooperation, informa-tion sharing and exchange techniques to be added tothe RbTbx will be instrumental for supporting theregional FMD interactions network. The developmentof these techniques was based on the way each stake-holder handles each data and information element. Forexample, proper system integration interfaces weresuggested for importing the data and informationelements produced by external entities [e.g. protectedareas maps, meteorological data and forest cover typemaps produced by the public administration (i2)]. Thelatter were to be accessed as “read-only” by the toolboxusers. Similar integration interfaces were designed foraccessing data and information elements managed byexternal proprietary information systems owned byusers of the RgTbx.

Providing external systems with easy access to thedata and information elements managed within theRgTbx was also a major concern. Therefore, thisresearch produced initial guidelines for the develop-ment of a Service Oriented Architecture for the RgTbx.Accordingly, the information services provided by theRgTbx modular components should be the basis for allinformation exchanges. These services should be direc-tly mapped to the data elements. They should rely ongeneral transaction standards for the forestry sector(e.g. Papinet 2011). Their technical documentationshould be easily accessible by the developers of otherexternal systems.

It further produced security policies to ensure thateach stakeholder has full control over the access to hisinformation, thus preserving data conf identiality.Specific access control modules were suggested to pa-rameterize the access permissions of all other usersafter a new data input by a stakeholder. Stakeholdersmay only access instances of information when bothits owners provide explicit access permission and theinformation element itself is relevant to his individualdecision processes. This security architecture was keyto sustain the conf idence of stakeholders on thetoolbox and thus to contribute to its use to support theirplanning processes namely the implementation and therecord of the management outcomes.


Furthermore, special attention was given to the dataand information elements editable by more than onestakeholder group. As an example, both ACHAR andthe forest service providers needed to Create, Update,Delete new equipment or equipment characteristics(i23). The future data and information governance mo-dels should include the identification of a unique stake-holder group responsible for each data element, thuscontributing to information consistency.

It further identified the potential users of each dataelement according to the integrated stakeholder analy-sis. In the case of data and information elements mana-ged by more than one stakeholder group, automaticworkflow procedures were recommended to ensureboth that the data instance could not be simultaneouslychanged by different stakeholders and that any inputor update is validated by all the stakeholders involvedbefore it became effective. For example, when theboundaries of a forest property are being updated, thesystem should trigger a procedure to prevent otherstakeholders from editing these boundaries. After-wards, the proposed boundaries update should beaccepted by other relevant stakeholders so that, forexample, it does not conflict with other existing boun-daries.

Lastly, data consolidation mechanisms were sugges-ted to enhance cooperation among stakeholders. As anexample, the information about forest interventionzones (i4) kept by ACHAR could be aggregated intomacro indicators (such as total forest area covered byZIFs). This will be help support forest strategy and re-gulation processes conducted by the national forestauthority with the participation of the forest fede-rations.

Discussion

The proposed approach for developing a regionalforest management planning toolbox (RgTbx) addressedthe need to acknowledge both the forest managementdecision processes by individual stakeholders and thecomplex regional interactions network. The RgTbxresulting from the application of this approach mayenhance the development of individual decision pro-cesses by providing access to innovative decisionsupport tools (e.g. models, methods and procedures).It may further support the stakeholders’ regionalinteraction network with adequate communication,cooperation, negotiation and information sharing

procedures and techniques. It will thus be influentialfor improving forest management planning at regionallevel.

The engagement of stakeholders in the design of theRgTbx was a major concern of this research. Therefore,participatory planning techniques were applied, par-ticularly during the process architecture workshops,according to the enterprise architecture methodologicalapproach. This approach has been used in Portugal fordesigning individual decision support systems, espe-cially for forest-based industries (e.g. Marques et al.,2010a,b) as well as for addressing interoperabilitybetween systems used to support the pulp and papersupply chain (Marques et al., 2010b; Marques et al.,2012). Yet, this research extended the EA methodologyto the architecture of a regional toolbox to be used ina multiple stakeholders’ context. The research challen-ges here the development of an approach to both ensurethe representativeness of the stakeholders involved inthe process and consolidate the results of the individualprocess workshops in an integrated stakeholders’analysis.

Thus, the proposed four-stage stakeholders’ engage-ment plan encompassed stakeholders’ selection, work-shops involving the stakeholders groups for individualdecision processes design, integrated stakeholdersanalysis, and the identification of the RgTbx compo-nents. Special attention was given to the stakeholders’selection stage in order to ensure representativenessand transparency. This stage is often a bottleneck to thedevelopment of the participatory planning processes.In order to overcome it, the project team must identifythe relevant players and promote their active involve-ment in the workshops. The cooperation with a key re-gional stakeholder (ACHAR) was instrumental for thesuccess of the engagement plan by this research.ACHAR promoted the project locally, made the con-tacts with the other relevant stakeholders and providedsupport to the research team moderator during theworkshops. The forest owners association was also themain promoter of the project outcomes and may con-tribute decisively to the implementation of the RgTbx.

The 13 stakeholders groups and the 22 entitiesdirectly involved proved to be representative of the re-gional forest planning context. They included the in-dustrial owners, the private forestland owners, theirassociation and federations, the national and regionaloffices of the forest authority, the forest investmentfund, the Chamusca municipality, the forest servicesproviders, the non-governmental environmental orga-


nizations and a research centre. The lack of involve-ment of the small-scale NIPF group was carefully con-sidered as it could pose as a limitation to the represen-tativeness of the results. Yet, recent studies on the be-havior of forest owners in Portugal (e.g. Batista andSantos, 2005; Novais and Canadas, 2010) provided theadequate rationale for their individual decisionprocesses. The project team representations were vali-dated afterwards with the forest association and thelocal communities. Nevertheless, the research dissemi-nation results could be enhanced by the active involve-ment of these forest owners.

The PA workshops were instrumental for documen-ting the stakeholders’ current decision processes aswell as for identifying their concerns and expectations.They further provided the opportunity to discuss withthe forestland owners the rationale of traditional andout-dated management practices and emphasize theadvantages of the adoption of new processes and newcomputerized-tools to support them.

The outcome of such interactive meetings washighly dependent on the moderator ability to clearlydefine the meeting objectives, establish a solid trustrelationship and frame their participation in the work-shop. The experience with Chamusca county provedthat these PA workshops should start with a brief pro-ject presentation, emphasizing its objectives and theexpected outcomes (specifying what is and what it isnot expected) to avoid misleading the participants andraising false expectations. This presentation shouldfurther provide a clear description of the method to beused during the sessions, including a reference to theterminology as well as to the graphical representationsto be obtained. These representations were built by themoderator on a paper board according to the answersof the stakeholders to the research questions. Theywere complemented by notes taken by the remainingmembers of the project team. Alternatively, classicalinterview methods were used whenever there wasevidence of lack of abstraction capacity by stakehol-ders. Yet these approaches did not encourage such anactive involvement of the participants.

The results of all the sessions were documented ona modeling tool (MS VISIO). The tool was not usedduring the workshops since it could interfere negativelywith the meeting dynamics. This tool was also used toautomate the generation of reports to be delivered af-terwards to the participants for content validation. Theworkshop with each stakeholder group encompassedat least two sessions. The first focused on the context

diagram while the second addressed the individualdecision processes, based on a draft proposed by theteam after the first session. In some cases, additionalsessions were conducted for validating the reports sin-ce most stakeholders did not reply when non-face-to-face methods were tried.

The individual and the integrated context diagramsprovided a good representation of the complexity ofthe regional interaction network. The analysis of thesediagrams further raised issues that can be properlyaddressed by forthcoming projects. The first issue wasrelated to the quantity, quality and the format of theRgTbx information to be accessed by the f irst-levelstakeholders. The content of current information flowsshould be screened during the development of theRgTbx data model to avoid superfluousness and lowquality.

The second issue was the reported scarce use ofcomputerized-tools by some stakeholders groups. Lackof training, small business size and specific businessrequirements not easily met by commercial systemswere among the explanations provided. Yet other fac-tors may be behind the potential resistance to compu-terized-tools utilization, such as ineffective dissemi-nation or inadequate forest extension services. Theacknowledgement of these factors may be instrumentalfor promoting the use of the RgTbx. The ongoingprojects involving industrial owners, forest ownersassociations, National Forest Authority and the forestresearch centre aiming at the development of forestmanagement decision support systems may suggestpotential collaborative approaches. Furthermore, thelist of required computerized-tools highlights futureresearch opportunities.

The third issue was the lack of cooperation practiceswithin the interaction network between stakeholdersbelonging to each group as well as between stakehol-ders at the same level in the context diagram. Theimplementation of the RgTbx should involve periodicmeetings and discussion forums to enhance communi-cation and cooperation between stakeholders.

An additional concern to be addressed by the RgTbximplementation is the need to reflect on the toolboxthe changes of stakeholders roles that result fromongoing initiatives for reorganizing the forestry sector.The integrated stakeholder analysis highlighted theprominent role of forestland owners in forest manage-ment planning in the region. However, this scenario islikely to evolve. The importance of other stakeholdersgroups is prone to increase in order to address emer-


gent regional and forest wide concerns (e.g. withwildfires). In this context, in the future forest associa-tions may play a more active role in forest managementplanning as they will become directly responsible formanaging the newly created intervention forest zones.The latter may also lead to a more active role in the de-velopment of commercialization strategies and cir-cuits. Recently, local communities and NGO haveacquired technical skills to participate and lead forestcertification schemas. This may contribute to increasetheir influence on management planning decisions aswell as on national and regional forest regulations.

The use of the RgTbx will trigger significant chan-ges to the Chamusca forest planning context. The useof the toolbox components will enhance the decisionprocesses by individual stakeholders. The optimizationmethods for evaluating alternative options and forcomparing the practices with other innovative solutionswill contribute to increase the efficiency and the effec-tiveness of forest management planning. These modelswill address other non-wood productions and forestservices as well as the economic and social sustainabi-lity of the regional forestry sector. Additionally, thetoolbox will facilitate the communication, cooperationand information sharing between the stakeholders,through the adoption of adequate system integrationand exchange mechanisms.

Nonetheless, the toolbox will foresee data accesscontrol features to be used by the data owners to regu-late the utilization of their information by others, thusovercoming the potential negative effects related withthe release of key business information consideredstrategic and confidential.

This research also underlined the need for rules go-verning the procedures and tools to support the net-work. Specifically, it suggested the development of anadequate organizational structure that could promotethe discussion about data governance models with thestakeholders. This new entity would be directly respon-sible for maintaining and updating the RgTbx and forproviding training and support to the users. Its gover-nance board should include representatives of the sta-keholders groups and the decision process within thisboard should reflect the relative contribution of eachstakeholder. Namely it should reflect the amount andrelevance of data and information elements kept by thestakeholder but accessible by other users. This aspectis of particular relevance since the RgTbx implementa-tion should include a schema for financing the produc-tion of certain information elements.

Conclusions

In this article, a regional toolbox was designed toaddress both the individual decision processes and thecomplex interactions networks between stakeholdersinvolved in forest management planning in a region.To the best of our knowledge, such a regional toolboxhas not been suggested in the forestry literature.

The stakeholders were actively involved in the tool-box design. The proposed stakeholders engagementplan extended the enterprise architecture methodologi-cal approach of Marques et al. (2010a,b) to apply par-ticipatory planning tools and techniques during in-teractive workshops to elicit decision processes withinforest management planning problems prevalent in aregion.

The identif ication of the RgTbx components wasbuilt upon these processes representations. Specif i-cally, it included a set of decision support tools (e.g.models, methods and procedures) already used or elsethat were required to address the stakeholders’ in-dividual decision processes, It further included a listof data and information elements needed to supportdecision-making. The RgTbx further included commu-nication, cooperation, information sharing and exchan-ge techniques aiming at supporting the regional inter-actions network.

The proposed stakeholders engagement plan fordesigning the regional toolbox was tested in the Cha-musca county, located in Central Portugal. The set ofstakeholders included 22 entities, divided among 13stakeholders groups, representing more than 900 peo-ple with direct interests in the region. The results showthat the proposed methodology may be used to providerepresentations of the individual decision processes.The representation of the information elements ex-changed and shared among stakeholders did provide aclear vision on the complex regional interactions net-works. Complementary, the integrated stakeholdersanalysis highlighted the key role of the forest owners,forest investment funds and regional offices of the fo-rest authority in the forest management planning fra-mework in the region.

Interactions and data exchange among these stake-holders were rare. Yet, all first level stakeholders wereinfluenced by the other stakeholders groups (e.g. theforest service providers, forest associations, forestbased-industries, Chamusca municipality and thenational office of the forest authority), acting mainlyas consultants or regulators. More than 20 other passive


stakeholders groups were indirectly engaged in forestmanagement planning in Chamusca, including forestresearch centers, local communities, sectorial associa-tions and NGO.

These results were instrumental for identifying 20decision support tools and 25 data and informationelements to be included on the RgTbx. 21% of thesetools and 48% of the data elements were already usedby the stakeholders.

The potential problem of data redundancy and in-consistency was characterized. This was influentialfor the suggestion of the communication, cooperation,information sharing and exchange techniques to beadded to the RgTbx. These techniques includedsystem integration interfaces and Service OrientedArchitecture features to regulate communicationbetween the RgTbx and other external systems. Bothdata governance models and access policies werefurther suggested for managing the data andinformation elements within the RgTbx in order toavoid redundancy, inconsistency and preserving dataconf identiality. Additionally, data consolidationmechanisms were suggested to enhance cooperationamong stakeholders.

Future research will divide the RgTbx into functio-nal sub-systems and provide their detailed functionalspecifications. It will follow the enterprise architectureapproach where these specifications emerge from theconfrontment between the presented Process Architec-ture and the subsequent Information Architecture,therefore assuring the alignment between the decisionprocesses and the IT function.

References

Ananda J, Herath G, 2003. Incorporating stakeholder valuesinto regional forest planning: a value function approach.Ecological Economics 45: 75-90.

Baptista FO, Santos R, 2005. Os proprietários florestais: re-sultados de un inquérito. Celta, Oeiras. 94 pp.

Borges JG, Hoganson HM, Rose DW, 1999. Combining adecomposition strategy with dynamic programming tosolve spatially constrained forest management schedulingproblems. Forest Science 45: 201-212.

Constantino M, Martins I, Borges JG, 2008. A new mixed-integer programming model for harvest scheduling subjectto maximum area restrictions. Operations Research 56(3):542-551.

Costa A, Oliveira AC, Vidas F, Borges JG, 2010. An approachto cork oak forest management planning in Southwestern Por-tugal. European Journal of Forest Research 129: 233-241.

De Steiguer J, Liberti L, Schuler A, Hansen B, 2003. Multi-

criteria decision models for forestry and natural resourcesmanagement: an annotated bibliography newtown Square,PA: US Department of Agriculture, Forest Service, North-eastern Research Station. 32 pp.

Díaz-Balteiro L, Romero C, 2008. Making forestry decisionswith multiple criteria: a review and an assessment. ForestEcology and Management 235: 3222-3241.

Falcão A, Borges JG, 2002. Combining random and syste-matic search heuristic procedures for solving spatiallyconstrained forest management scheduling models. ForestScience 48(3): 608-621.

Falcão A, Próspero dos Santos M, Borges JG, 2006. A real-time visualization tool for forest ecosystem managementdecision support. Computer and Electronics in Agriculture53: 3-12.

Ferreira L, Constantino M, Borges JG, 2011. Optimizing themanagement of Pinus pinaster stand, under risk of fire.An application in Portugal. Annals of Operational Re-search.

Forsell N, Wikström P, García F, Sabbadin R, Blennow K,Eriksson LO, 2009. Management of the risk of wind da-mage in forestry: a graph-based Markov decision processapproach. Annals of Operations Research.

García-Gonzalo J, Pukkala T, Borges JG, 2011. Integratingfire risk in stand management scheduling. An applicationto Maritime pine stands in Portugal. Annals of OperationalResearch.

Grimble R, Wellard K, 1997. Stakeholder methodologies innatural resource management: a review of principles, con-texts, experiences and opportunities. Agricultural systems55(2): 173-193.

Hjorts CN, 2004. Enhancing public participation in naturalresource management using Soft OR – an application ofstrategic option development and analysis in tactical forestplanning. European Journal of Operational Research. 152:667-683.

Kangas A, Kangas J, Kurttila M, 2008. Decision support for forest management (Gadow K, Pukkala T, Tomé M, eds). Managing Forest Ecosystems Vol 16. Springer.224 pp.

Korpela J, Sierilä P, Tuominen M, 2001. A GDSS-basedapproach to the strategy analysis of forest industries. In:Proceedings of the 34th Annual Hawaii International Con-ference on System Sciences (HICSS-34), IEEE ComputerSociety.

Kurtilla M, Pukkala T, 2003. Combining holding-level eco-nomic goals with spatial landscape-level goals in the planningof multiple ownership forestry. Landscape Ecology 18:529-541.

Marques AF, Borges JG, Sousa P, Pinho AM, 2010a. Anenterprise architecture approach to forest managementsupport systems design. An application to pulpwoodsupply management in Portugal. European Journal ofForest Research 130(6): 935-948.

Marques AF, Borges JG, Sousa P, Fonseca M, García R,Batista G, 2010b. Applying enterprise architecture to thedesign of the integrated forest products supply chainmanagement system (Varajão et al., eds). Proceedings of


CENTERIS 2010, Springer-Verlag Berlin Heidelberg. pp:32-40.

Marques AF, Borges JG, Sousa P, Fonseca M, Gonçalves J,Oliveira J, 2012. Chapter 1: An enterprise architectureapproach for designing an integrated wood supply mana-gement system. In: Organizational integration of enterpri-se systems and resources: advancements and applications(Varajão et al., eds). IGI Global.

Martins H, Borges JG, 2007. Addressing collaborativeplanning methods and tools in forest management. ForestEcology and Management 248(1-2): 107-118.

Novais A, Canadas MJ, 2010. Understanding the manage-ment logic of private forest owners: a new approach.Forest Policy and Economics 12: 173-180.

Nieuwenhuis M, Tiernan D, 2005. The impact of the intro-duction of sustainable forest management objectives onthe optimization of PC-based forest-level harvest schedu-les. Forest Policy and Economics 7(4): 689-701.

Nordstrom E-M, Romero C, Eriksson LO, Öhman K, 2009.Aggregation of preferences in participatory forestplanning with multiple criteria: an application to the urbanforest in Lycksele, Sweden. Canadian Journal of ForestResearch-Revue Canadienne de Recherche Forestiere 39:1979-1992.

Palahí M, Pukkala T, 2003. Optimizing the management ofScots pine (Pinus sylvestris L.) stands in Spain based on in-dividual-tree models. Annals of Forest Science 60: 105-114.

Papinet, 2011. Global standard for the paper and forestproducts. PapiNet ISS tool user guide v2r20. [internet si-te]. Available at www.papinet.org [Cited 20 Jun 2011].

PMI, 2011. A guide to the project management body ofknowledge (PMBOK Guide). PMI Standards Committee,Project Management Institute. [Internet site]. Availableat www.pmi.org [Cited 20 Jun 2011].

Pukkala T, Heinonen T, Kurttila M, 2009. An application ofa reduced cost approach to spatial forest planning. ForestScience 55(1): 13-22.

Purnomo H, Mendonza G, Prabhu R, Yasmi Y, 2005.Developing multi-stakeholder forest management scena-rios: a multi-agent system simulation approach appliedin Indonesia. Forest Policy and Economics 7: 475-491.

Ribeiro RP, Borges JG, Pereira C, Sousa P, Lé J, 2005.Designing an integrated forest planning system for theforest industry: an application in Portugal. In: Proceedingsof the Symposium on Systems Analysis in Forest Resour-ces, Stevenson, WA, USA.

Schmoldt D, Kangas J, Mendoza G, 2001. Basic princi-ples of decision making in natural resources and theenvironment. In: The analytic hierarchy process in natu-ral resource and environmental decision making(Schmoldt D et al., eds). Kluwer Academic Publishers.pp: 1-13.

Sheppard S, Meitner M, 2005. Using multi-criteria analysisand visualization for sustainable forest management plan-ning with stakeholder groups. Forest Ecology and Mana-gement 207: 171-187.

Sousa P, Pereira C, 2005. Enterprise architecture: businessand IT alignment. In: Proceedings of the Symposium onApplied Computing Archive, Santa Fe, New Mexico.

Sousa P, Pereira C, Alves Marques J, 2005. Enterprise archi-tecture alignment heuristics. Microsoft Enterprise Archi-tectural Journal.

Spewak S, Hill S, 1992. Enterprise architecture planning:developing a blueprint for data, applications and techno-logy. Wiley-QED Publication.

Tecle A, Shrestha B, Ducksteins L, 1998. A multiobjectivedecision support system for multiresource forest manage-ment. Group Decision and Negotiation 7: 23-40.

Toth SF, McDill ME, 2008. Promoting large, compact matureforest patches in harvest scheduling models. Environmen-tal Modeling and Assessment 13(1): 1-15.

Vainio P, Tokola T, Palander T, Kangas A, 2009. A GIS-basedstand management system for estimating local energywood supplies. Biomass Bioenergy 33: 1278-1288.


/ColorImageDict > /JPEG2000ColorACSImageDict > /JPEG2000ColorImageDict > /AntiAliasGrayImages false /CropGrayImages true /GrayImageMinResolution 300 /GrayImageMinResolutionPolicy /OK /DownsampleGrayImages true /GrayImageDownsampleType /Bicubic /GrayImageResolution 300 /GrayImageDepth -1 /GrayImageMinDownsampleDepth 2 /GrayImageDownsampleThreshold 1.50000 /EncodeGrayImages true /GrayImageFilter /DCTEncode /AutoFilterGrayImages true /GrayImageAutoFilterStrategy /JPEG /GrayACSImageDict > /GrayImageDict > /JPEG2000GrayACSImageDict > /JPEG2000GrayImageDict > /AntiAliasMonoImages false /CropMonoImages true /MonoImageMinResolution 1200 /MonoImageMinResolutionPolicy /OK /DownsampleMonoImages true /MonoImageDownsampleType /Bicubic /MonoImageResolution 1200 /MonoImageDepth -1 /MonoImageDownsampleThreshold 1.50000 /EncodeMonoImages true /MonoImageFilter /CCITTFaxEncode /MonoImageDict > /AllowPSXObjects false /CheckCompliance [ /None ] /PDFX1aCheck false /PDFX3Check false /PDFXCompliantPDFOnly false /PDFXNoTrimBoxError true /PDFXTrimBoxToMediaBoxOffset [ 0.00000 0.00000 0.00000 0.00000 ] /PDFXSetBleedBoxToMediaBox true /PDFXBleedBoxToTrimBoxOffset [ 0.00000 0.00000 0.00000 0.00000 ] /PDFXOutputIntentProfile () /PDFXOutputConditionIdentifier () /PDFXOutputCondition () /PDFXRegistryName () /PDFXTrapped /False

/CreateJDFFile false /Description > /Namespace [ (Adobe) (Common) (1.0) ] /OtherNamespaces [ > /FormElements false /GenerateStructure false /IncludeBookmarks false /IncludeHyperlinks false /IncludeInteractive false /IncludeLayers false /IncludeProfiles false /MultimediaHandling /UseObjectSettings /Namespace [ (Adobe) (CreativeSuite) (2.0) ] /PDFXOutputIntentProfileSelector /DocumentCMYK /PreserveEditing true /UntaggedCMYKHandling /LeaveUntagged /UntaggedRGBHandling /UseDocumentProfile /UseDocumentBleed false >> ]>> setdistillerparams> setpagedevice

340-358 3120 a participatoryenterprise architecture methodology firstly presented by spewak and hill...

Documents