Copyright 2004, Society of Petroleum Engineers Inc. This paper was prepared for presentation at The Seventh SPE International Conference on Health, Safety, and Environment in Oil and Gas Exploration and Production held in Calgary, Alberta, Canada, 29–31 March 2004. This paper was selected for presentation by an SPE Program Committee following review of information contained in a proposal submitted by the author(s). Contents of the paper, as presented, have not been reviewed by the Society of Petroleum Engineers and are subject to correction by the author(s). The material, as presented, does not necessarily reflect any position of the Society of Petroleum Engineers, its officers, or members. Papers presented at SPE meetings are subject to publication review by Editorial Committees of the Society of Petroleum Engineers. Electronic reproduction, distribution, or storage of any part of this paper for commercial purposes without the written consent of the Society of Petroleum Engineers is prohibited. Permission to reproduce in print is restricted to a proposal of not more than 300 words; illustrations may not be copied. The proposal must contain conspicuous acknowledgment of where and by whom the paper was presented. Write Librarian, SPE, P.O. Box 833836, Richardson, TX 75083-3836, U.S.A., fax 01-972-952-9435.

Acknowledgements We would like to thank Mr. Alan Hammer QAC Manager, Operations Coflexip Stena Offshore. Mr. Stacy P. Payne, Manager International HS&E, Global Marine. Dr. Mark Fleming, Chartered Psychologist The Keil Centre. They all shared extensively from their vast knowledge and experience, and provided both encouragement and guidance at the outset of our project. We truly believe that their valuable input so far has enabled us to avoid the worst pitfalls in such programs. Proposal This paper describes the introduction of a management tool for behavior modification implemented in Statoils E&P operations over the past two years. A large number of executives, managers and supervisors1 have been trained, in small groups, to use a structured methodology for work site interaction with staff aimed at reducing at risk behavior in our plants and facilities. Our training program have been highly standardized and at the same time focused on stimulation of dialogue and creation of consensus amongst the participants. Our Open Safety Dialogue (OSD) introduction program has been conducted in the shortest possible time span. Each supervisor has been put under mild pressure to apply the OSD methodology at completion of her/his training. We will briefly describe the OSD methodology, our training program and the tools and techniques we have developed and applied to achieve our objectives.

1 Supervisor is used as a generic term for executives, managers, supervisors and foremen in the rest of the text.

The main objective is to share our experiences with respect to: Culture conflicts encountered How to ensure usage and compliance Reporting and statistics Managerial style elements Opposing forces Employee reactions The requirements and tools for continued pressure and long term commitment Surprises and rewards We will also present our plans for OSD follow up activities and our related peer program; which are now in progress. OSD methodology As in most behavioral safety programs we apply elements of scientific psychology, labeled as “coaching”, in our program. The central element of OSD is that two supervisors invite employees who are at worki to engage in an open, honest and friendly dialogue aimed at having the employees themselves realize and express their at risk behaviors, as well as the potentional consequences to themselves, with respect to ongoing work at the actual work sites. It is mandatory for the supervisors in an OSD to demonstrate a 100 % no blame attitude. This in recognition of the fact that a significant percentage of individual at risk behaviors are unconscious, and to avoid provoking a counterproductive defensive reaction from the employees. The dialogue is closed with a formal agreement of behavioral change, when such changes are required. These agreements are followed up. Supervisors will not use OSD in situations where they observe gross violations of safety rules and regulations that qualify directly for disciplinary action. Dupont have, in a recent report to Statoilr1, stated: “It is clearly recognized that the OSD program is a copy of previous versions of Dupont’s “STOP for Supervisors” program….” Even though our program and STOP, and a series of similar programs, have the same basic foundation in psychology we beg to differ from this “is a copy of” assessment We have sourced our methodology from several different existing programs (some of them recognized in industry as based on Dupont’s HSE programs; such as STOP, START, ABS, ASA etc.) as well as published papers and reports.

SPE 86625

Open Safety Dialogue (OSD), Dialogue, Delivery and Commitment Terje Løvby and Tor Alrik Dahl, Statoil ASA

2 SPE 86625

It was however imperative to our objective to tailor our final solution both to complement existing corporate HSE standards and systems, as well as to be adapted to the Norwegian work culture. We feel that this is clearly acknowledged in Dupont’s report where it is stated. “It is, however, recognized that as an internally developed program there are a number of variances, both in terms of concept and implementation when compared with Dupont best practice”. (We will identify some of our variances from Dupont “best practice” throughout this paper.) Training program Our formal training is conducted as a one day course at sites / plants and offshore platforms with a small group of supervisors per instructor. The theory part addresses both the deficiencies of previously applied HSE supervisory techniques and the reasons why we are convinced that correctly applied “coaching” (i.e. OSD) will have the desired effect, over time. Our structured dialogue methodology is then introduced step by step. Each step is justified by referring both back to the participant’s own knowledge and experiences and psychology explained in previous sessions. The theoretical part of the training is concluded by role-plays based on photos of realistic work situations selected from the course participant’s own areas of responsibility. The practical part of the course requires the participants to engage in real life dialogues based on what they have learned so far. Each of the participants typically gets the benefit of participating in 2 – 4 OSDs, under instructor supervision and guidance. All the course participants are invited to establish their own personal target for how many OSD they will try to participate in during the coming year, at course closure. They do this by filling in an agreement with the CEO of our company at the end of the course. (Fig.1). (The instructor copies these for data recording, and the participants keep their original agreement as a course diploma).

Fig. 1 Personal OSD ambition agreement

The OSD training is complemented by a documentation package. This is only distributed to course participants. It was not intended, nor developed, as self-study documentation but as a reminder and supplement to what was explained during the course. Our experiences The project team who initiated this program in late 2001 highlighted that the only true measure of success for a behavioral modification program would be a continuous and significant reduction in LTA and injury frequency over a long period of time (years). It was also realized that a broad and significant change in safety culture was a critical success criteria. And that such changes require both “top-down” and “across the board” focus and involvement, as well as time. The project teams postulate was that it would take at least 3 – 5 years of dedicated work before true and significant results would materialize. It is therefore far to early in the project to assess and report measurable (“true”) achievements. We have however already observed several interesting issues we will like to share with our readers. Although most of these observations ( )2 are of empirical nature, we do believe we are in a unique position to state some of these as “facts” because a low number of instructorsii have trained 3700 supervisors in small groups over a period of less than two years. The same instructors have also coached a significant number of the same supervisors in use of OSD in their work environment. (This coaching typically follows 8 –10 months after their initial OSD training). As we have trained and coached a large number of supervisors from our main contractor companies in addition to our own employees, representing a multitude of professional backgrounds, nationalities, and corporate cultures, we truly feel that our observations are representative not only for different companies but also for different nationalitiesiii. Our observations and experiences are in other words of a multi nationality and a multi company nature and therefore not in any regard only limited to our own company. Culture conflicts encountered Norwegian organizational culture r2 The overall Norwegian business organization culture and managerial styles have been developed and refined over the last 5 – 6 decades. A society founded on broadly accepted social democratic principles has had a strong impact on the way we organize managerial structures and workforce relations in the Norwegian industry.

2 We have used this marker ( ) throughout the rest of our text to identify our main observation- / solution statements.

SPE 86625 3

The supervisors and employees in Norway are considered equal, particularly with respect to matters that directly impact the workforce. Unions and collaborative bodies within the organization are considered to be viable channels for appeal of managerial decisions. The supervisors true authority is to a large extend dependent on the individual’s professional skills and contact network and not so much on rank and position. Supervisors may as a result of this be questioned, and even interrogated, from multiple levels and directions in the organization, even on minor issues. Direct orders and demand for loyal adherence to procedures, standards, and (to a degree) even laws, are “out of fashion” in Norway as involvement and consensus are the overall recommended managerial styles in our country.

- - - We also have, to a large extent, abandoned the use of disciplinary actions in enforcing good safety regulations and procedures.

It is generally believed in Norway that Norwegians are able to practice open and candid communication techniques in all matters at all levels (as long as it does not involve very personal feelings and relations).

We were therefore greatly surprised to find that most Norwegian supervisors reverted to the old “HSE inspector and fault finding technique” when their objective was to guide their employees, in a positive manner, to assess their own at risk behavior and the potentional worse personal impact, if an incident happened.

The much praised openness and informality in Norway do not ensure that supervisors are visible at the work sites. Steadily increasing parts of their time are spent as “desk jockeys”, or in meetings. Even when the message from the company’s executive board, combined with the OSD program, are that more visible leadership is required supervisors themselves tend to “ease” back into the same office work most of them claim to detest. We know, from direct communication, that this lack of visible leadership in our plants is both observed and of concern to our CEO. How to ensure usage and compliance Statoil has, as most other companies, experienced previous well-intended and important HSE initiatives being received with enthusiasm and motivation both by management, unions and the work force before they fizzle into a halt for no obvious reason. We also had some knowledge about problems in other organizations behavior modification projects, where each supervisor was assigned a quota to perform within a set

timeframe. (i.e. projects which turned into pure “number games”) We knew before this project commenced that most of our supervisors felt that they already had their “plate” more than filled with priority tasks. We therefore had to secure across the organization priority and dedication to the OSD challenge, while we at the same time did not have available resources to conduct ongoing on site monitoring of progress and performance.

It was decided that we would require the supervisors to report each conducted dialogue. In order to counteract negative reactions, funded on previous experiences we had to avoid some of the traps we, and others, had fallen into in earlier projects:

• We require only confirmation of an OSD being conducted

• We made the reporting as efficient as possible • We do not require any reporting of observations iv • We measure performance per organizational unit

only, not by individual supervisor v • We enable extensive reporting for those who so

desire. • We prohibit identification of workforce participants • We established a long term program of coaching and

verification We experienced a lot of bickering and moaning from supervisors about “yet another reporting system” at the outset. The claim was that they would do the dialogues regardless of reporting or not, and that the quality of the OSDs would be better if they did not have to worry about having to report them. We therefore waited patiently and did not enforce reporting initially. Each unit was questioned about their progress after approx. 3 months, and they admitted, even without probing, that progress had not been particularly good. Senior Management then demanded active reporting; and the worst objections to this were silenced.

We have been extremely surprised to discover that supervisor teams in our production facilities and even groups of supervisors on individual shifts have developed their own standards of OSD methodology, which in some cases even are in direct conflict with the detailed and justified instructions which are given during our structured formal training or in the documentation. And what’s even worse; they claim that their approach is based on and consistent with the same sources. We have, based on this observation, decided that OSD instructors periodically will visit all plants/shifts and perform coaching of individual supervisors by participating in OSDs on one-to-one basis.

4 SPE 86625

Reporting and statistics We have verified through targeted testing that extraordinary

tasks like OSDs will not get done in large organizations unless a reporting is required and controlled. The only mandatory reporting of an OSD consists of date, time, plant, area (in plant) and the names of the two supervisors who initiated the dialogue. The reporting is done through a Lotus Notes database form (Fig 2) where the few mandatory fields are clearly highlighted. The reporting is made as efficient as possible, and we have proved through tests that even a first time user needs less than 30 seconds to record their first OSD. We have facilitated extensive reporting features and follow up assistance for supervisors who want to utilize these. We do also allow for manual recording and reporting through assistants (secretaries, planners etc.) if required. The only thing that is not facilitated, or permitted, is identification of the employee(s) the supervisors talked with. We have received much praise for the simplicity of our Lotus Notes reporting solution.

Fig. 2 Database reporting form Our company has other systems where the business units in Statoil report their manpower employed per period, in order to facilitate multiple types of frequency reporting.

This combination permitted us to develop a management reporting scheme where we on a monthly basis present the achieved OSD factor for each plant/org. unit. This shows the number of OSDs factorized over the number of man-years employed in the same period. (Fig. 3) We don’t have any other reporting scheme for the project. The OSD factor enables the executives to compare OSD

activity between plants/org. units regardless of size as well as to establish the same target for any level of our organization. We have established a factor of 2 as a target for the first year in full operation (2003) and do not report any achievement higher than this target. (To circumvent any “best guy in class” competition between units). The OSD factor will be/is included in the performance- / bonus contracts for plant managers. It’s our overall plan to increase this target in a controlled fashion on an annual basis until we reach a company wide OSD factor of at least 7.

10.11.2003 OSD 2003 OSD TEAM E&P N HSE 1

0

0,5

1

1,5

2

2,5

Platf

orm

APl

atfor

m B

Platf

orm

CPl

atfor

m D

Platf

orm

EPl

atfor

m F

Platf

orm

GPl

atfor

m H

Platf

orm

I fro

m 15

.05.0

3

Platf

orm

J fro

m 0

1.02

.03

Platf

orm

KPl

atfor

m L

Platf

orm

M fr

om 1

0.03.

03

Platf

orm

N fr

om 2

4.03.

03Pl

atfor

m O

Platf

orm

PPl

atfor

m Q

Proj

ect A

from

01.0

4.03

Platf

orm

R fr

om 2

4.06.

03Pl

atfor

m S

Plan

t A fr

om 0

1.05.

03

Plan

t BOn

shor

e bas

e(s)

Proj

ect B

from

01.

06.0

3

OSD factor 01.11.2003 for Offshore platforms and onshore plants E&P - NG

Factor = No of OSDs per unit/plant divided by man-years employed (same period ). Offshore man-year = 1615 hours Onshore man-year = 1820 hours.

No. of OSDs

The score limit for year 2003 is set to 2 (meaning “2 and over”) to reduce the risk of stimulating a pure ”numbers competition” .

47143153 205104 340 238 598 320 449 237 475 139 378 154 3979551203398540800117612611191

Fig. 3 Monthly executive report, OSD factor The project team has started assessing which type of statistical presentations may be useful in the future. Although it is far to early to interpret any of this as significant measures we believe that we already see some initial confirmation of the fact that you have to achieve a minimum “critical mass” of OSDs before you can expect to see stable and positive results from such efforts (Fig 4).

0

5

10

15

20

25

30

jan.03 feb.03 mar.03 apr.03 mai.03 jun.03 jul.03 aug.03 sep.030

0,2

0,4

0,6

0,8

1

1,2

1,4

1,6

1,8

2

Injury ferquency OSD Factor

Injury frequency - OSD Factor

Platform A Platform B

Fig. 4 A very preliminary observation of results

SPE 86625 5

Managerial style elements Our executives and senior managers lecture about safety in

large groups in the correct manner; - they know very well what to say. But to demonstrate and verify the same gospel by “walking around” and stimulate the work force directly has proved to be quite another and far more challenging task. This in spite of the fact that the same executives, during OSD training, agree that the personal positive dialogue, practiced at work sites may be more effective, also in a wider perspective. We feel that the paradox most executives face in this regard is that there always are so many major business issues and tasks at hand that a few brief and short (although extremely important) dialogues about worksite safety all times runs a high risk of being slightly delayed and postponed. Then, as all executives work periods and site visit programs have a tendency to be overloaded, it ends up with a repetitive personal promise of doing a better job in this area at the next opportunity. We do however know that our executives are aware of this situation and are actively doing their best to facilitate opportunities for OSDs in their schedules during site visits.

Supervisors who have an extensive professional background within their own area of responsibility seem to be the ones who struggle the most to make these dialogues about their employees personal safety natural. This in spite of the fact that they often have both good and close personal relationships with their colleagues. They revert very quickly into a dialogue about purely technical topics or the old “finger pointing” as soon as they sense that the communication gets a little “strained”.

We have in all our OSD training and coaching highlighted the fact that it would generate a high risk for overall program failure if we try to stimulate OSD activities by introducing quotas per supervisor. Dialogues of this nature is not something all supervisors, as well as employees, will feel comfortable with initially. To neglect important this fact will only stimulate counterproductive reactions and reporting. We feel that we have successfully argued this case to our top executives, but are still engaged in an ongoing struggle to prevent some representatives of our middle management from applying such negative performance measures within their area of responsibility.

Quite a number of the supervisors see dialogues such as this

as a strange sort of competition between themselves and their crewmembers. As a result they engage in pure power plays where they have to find something at fault (“the auditor syndrome”) even when there is nothing truly wrong in the situation.

The OSD represents an extremely simple, and easy to understand and justify, dialogue format. In spite of this it appears to be extremely difficult to learn and practice by seasoned supervisors.

At the same time when we observe this hard struggle to learn and be confident in OSD practice by our supervisors

during the practical part of OSD training and coaching we are surprised by requests for shorter training, “light versions” and other “quick fix” solutions with respect to the formal training from the very same supervisors who struggles the most.

Opposing forces

We have for a number of years worked diligently to make individual workers truly realize that they themselves are the persons most responsible (“in charge”) for their own personal safety. It appears now that we, as an undesired side effect of this effort, may have reduced the feeling of direct responsibility for the same issues on the supervisor’s part.

We have also experienced company wide initiatives for cost

cutting and manpower reductions like all the other oil companies in the world. This has required resources for project work, process changes, changes in the supervisor’s authority and responsibilities and empowerment of workforce versus supervision. Introduction of new and powerful corporate IT systems, which at the same time are more demanding on the end users, are also tying the supervisors more than to their offices and PCs. These developments have reduced the available time for supervisors to visit worksites and engage themselves in ongoing work activities for their crews. We expected this to have a significant negative impact on our project, but this has failed to materialize. On contrary it seems that the new OSD initiative where supervisors more frequently visit employees in the plant area and focus on personal safety and display honest concern for each individual, to a degree have reduced the anxiety which could have been caused by the corporate cost cutting ambitions. We have also recorded frequent feedback, from both parties, to the fact that OSDs are welcomed as a missed closer level of ongoing contact between our employees and their supervisors.

Executives, managers and supervisors strongly agree during

OSD training that our failure to use disciplinary action in a correct fashion as a tool to enforce important safety rules and procedures is a major obstacle to the development of a good safety culture in our company. We have an established practice of handling a few extreme safety violations through disciplinary actions, but at the same time have neglected to pursue people who repetitively violate mandatory safety precautions and procedures through an orderly application of our disciplinary procedure. It is “a known fact” that we, as a result, have people in our organization who choose to behave in a fashion with respect to safety where they are a real danger both to themselves as well as their colleagues. This has also been identified in the Dupont report mentioned earlierr1 and is now an important action item in our plan for 2004.

A significant number of the course participants are very active during training, subscribe to all the evaluations and arguments presented, are enthusiastic about the potentional of the OSD methodology and give the training very high scores

6 SPE 86625

on the course evaluation. (Overall average approx 5,2. On a 1 – 6 scale). Even so, a significant number of the supervisors who give the course top score only commit themselves to carry out 2 – 5 OSDs per year at course closure. Employee reactions

The work force is in general very positive to the OSD experiences they have encountered, partly because they have truly missed interactions with their supervisors (visits) in the workplace. The union’s safety delegates are voicing a joint strong opinion stating that OSD have been a significant and very positive new approach to safety work in our company. We have however, through the “grapevine”, received some, hopefully isolated, observations of rather disastrous dialogues. Observations of this nature which have reached the project team members in person, seems to originate with disloyal supervisors who choose to display their discontent and disagreement with the OSD initiative to subordinates instead of addressing and resolving this with their superiors. The requirements and tools for continued pressure and long term commitment The most important factor, at this stage in the program, in order to secure long-term commitment and success for our behavioral safety program, will be to demonstrate solid and personal top management commitment and engagement. This is the area where we have struggled the most in the last two years. Our executives are both dedicated to the program and actively demanding supervisor performance and OSD reporting. They monitor the OSD factor on a monthly basis and both challenge and praise individual managers and units through direct communication. We know that our executives are quite good at doing OSDs, when they manage to find the opportunity and time to perform them. And it is greatly appreciated by the work force when they observe members of the executive board and company presidents through OSDs both being interested in their work and cautioning them in a friendly manner to take care of themselves confirming the corporate safety credo “We do always have the required time to work in a safe and controlled manner”

The paradox is that with all these immensely important factors our executives really struggle to find the time and opportunity to carry out a few short dialogues with such potentional and impact, even when they visit plants and sites. This dilemma have now the required executive attention and our OSD project team have been requested both to continuously challenge them on their performance in this regard as well as to facilitate arrangements where they actively can demonstrate their OSD dedication. We are therefore confident that we will see a significant positive development

with respect to visible behavioral change commitment from key members of our corporate executive committee. Surprises and rewards

The biggest surprise has been that we completely under-estimated the training requirements. We settled for an one-day formal course to secure that even the busiest supervisors could find the time to attend such limited training and because we thought it would be more than sufficient time to learn to use and practice a very simple and productive communication technique. We train the supervisors to carry out a friendly dialogue with people they know, which should last no more than 5 minutes, through the use of four basic questions (which we even provide on a “memory jogger” card). As basis for the practical training we establish a joint understanding as to the basis for OSD as a powerful tool to improve safety. Understandings the majority of our course participants subscribes to, and is enthusiastic about. In spite of this positive foundation we have experienced that our supervisors struggle, and require far more training than we expected when we developed the program. We are convinced that this to a large degree is caused by the fact that we previously have trained our supervisors so well in pure faultfinding and “finger pointing” methods. Methods they have practiced for years. We completely overlooked this need for “clean out training”, and have to rectify this now, when the initial OSD training should be approaching an end.

Although we knew that our supervisors over time have been turned more and more into office workers we did not realize to which degree this had removed them from hands on knowledge about what actually goes on at the work sites. Quite a few of our supervisors have had major “eye openers” after they have started practicing OSD. We are also recording a very high degree of appreciation from the work force confirming that they have actually missed the contact with their supervisors out in the field (at their work sites).

An observation by the OSD instructors, which have increased in magnitude as the program have progressed, is the lack of overall consistency in our company’s and employees safety focus. While we consider ourselves quite good in this regard in our plants and on offshore installations we often talk about this being “carried over” to the rest of our organization as well as our employees private sphere (homes etc.) “24 hours HSE focus”. As OSD instructors, traveling around in the different environments and also talking to supervisors and staff about their off hours safety focus, we collect the most incredible observations and stories showing that such safety carry over effects often are only marginal and even nonexistent. Gross at risk behaviors, that would have been addressed immediately at our plants, are often neglected when they occur in and around our office buildings, even by visiting plant supervisors. Plant workers and supervisors, who are recognized as very safety conscious at work, tell numerous stories about near misses at home due to their own at risk behavior, within areas they would never fail in at work etc.

SPE 86625 7

And a number of our supervisors, who know that they should “lead by example” at work, admits they tend to do otherwise when they are at home with their own family. “Safety is not something which is transmitted, it has to be focused at and worked on at all times, everywhere”

One of our immediate and great rewards have been to be

present during OSDs where the dialogue technique used correctly have revealed unseen but significant at risk behaviors with a potentional for severe injuries; and where the employees are truly grateful for having been made aware of the dangers.

Another reward has been meeting supervisors who states that this is the best professional experience and leadership training they have encountered in their professional lives.

We also consider the support and engagement from our CEO and corporate executive committee members, and their commitment to long-term focus and push for this project, as a reward for the project team. As said above, we would like to see this extended to more direct OSD activities on their part as well. We do however expect to see a positive and significant change in this in the near future. Future plans and activities OSD follow up activities We will increase our OSD coaching in order to compensate both the requirements for more practical training and to counteract the tendency to develop sub standards locally in our company. We will develop and conduct opinion polls aimed at collecting better information as to how OSD is experienced both by the workforce and the supervisors. This will be anonymous polls where we only aim to be able to identify whether the respondent is a supervisor or employee, and at which plant the respondent are employed. We are already in the process of developing an OSD Newsletter. This is aimed at easing ongoing communication with our supervisors regarding OSD, and at the same time a vehicle to enable communication about our company standard as to the methodology and practice of OSD. We aim to start a series of motivation efforts. The first objective is to communicate directly by letter (in a positive manner!) with supervisors who might believe they have been completely forgotten about how we might help them to achieve their personal objectives with regard to the number of OSDs they aimed to do at the end of training. The next stage will be to give some sort of recognition to plants and departments who have achieved the annual OSD factor and even to give the individuals of these units some sort of recognition. The CEO has requested that a parallel program is developed to train and motivate all senior support personnel (engineers,

purchasers, corporate management, etc.) who through their actions, or lack of such, may impact the safety of others. As an important part of this program version we will also increase our focus at office and “off hours” safety for all company employees and environments. We are also looking into ways and means to introduce some sort of competition into the OSD program. As we, at all costs, will avoid stimulating a situation where we focus on the number of OSDs per individual supervisor our current thoughts are on rewarding “The good OSD” or “The OSD with the best experience transfer potentional” etc. Peer program The project proposal for OSD underscored that this is a top – down program, and stated that it had to be complemented by a peer program as a crucial part of an overall behavioral safety program. The project plan proposed that the peer program was phased in approximately 1,5 year after the start OSD training. Such a delay was required both in order to secure that OSD would be well established before the peer program started, and to avoid depleting OSD for crucial resources before it reached the half way mark. The peer programs objectives include reduction of communication barriers between colleagues with respect to dialogue about at risk behaviors, and openness about private matters that may impact individual’s focus and safety. It also include awareness training as to injury consequence assessment by documented experiences and highlights the importance of both taking care of each other and staying focused so as to ensure that all individuals are going home from work in equal or better shape as when they came. Statoils safe behavior peer program is now ongoing. According to current plans it will engage more than 10,000 company and contract employees over a period of 2 years participating in large conventions of 2 days duration complemented by a 4 year program of follow up activities. The supervisors who have completed OSD training are important role models and resources in this peer program There is tight collaboration between the two programs but because of the unexpected and extended demand for OSD training a separate project group is in charge our safe behavior peer program Afterthought

Old habits die hard,

regardless of whether it’s managerial styles or at risk behaviors at the work sites.

– In order to have success in a behavioral safety program

both must be changed

8 SPE 86625

References r1. HSE MANAGEMENT EVALUATION for Statoil ASA, Stavanger; Norway, March 2003; (Internal report made for Statoil by Dupont) r2. Culture and HSE Management, Sigurd K. Berg, Aker Offshore Partner; SPE 73993, Society of Petroleum Engineers Inc. 2002. Endnotes i This is one of the areas where we deviate from the way we have seen Dupont’s STOP program being practiced. We have OSD dialogues with people at work, regardless of whether we observe anything wrong or not. This defuses any anxiety on the part of the workforce regarding supervisors interrupting them for such dialogues when they are working. ii A very small number of instructors do all the OSD training in Statoil. The objective is to secure that we present a consistent message and a structured methodology with standard elements. We do not permit “cascade training” in the OSD program. (This is another area where we deviate from Dupont’s standard) iii We have trained Norwegians, Danes, Swedes, Germans, Dutchmen, Portuguese, Spaniards, Americans, Englishmen, Scots and Chinese from more than 30 different companies. iv This is an area were we clearly deviate from Dupont’s “best practice”. We do not collect STOP cards and process them by HSE committees etc. We are of the opinion that the intended result of the dialogue (that the individual workers realize that they should focus on eliminating their own at risk behaviors) far outweighs any potentional for organizational learning through committees. Another reason for this policy is that we feel that a practice of handling such reports in HSE committees will pose a high risk of violation of our “no name – no blame” ambition. We have another incident reporting system (Synergy) that aim to facilitate important experience transfer and we do not feel that it’s advisable to have multiple reporting and administrative systems with the same objectives. v We have established a measure of performance where we compute the number of OSDs per man-year employed for the same period. For our first year of “full operation” we set the target for all operational units to factor 2. It is our ambition to increase this target over time until we reach an OSD factor of at least 7 for all units. The individual ambitions of each supervisor are only known to themselves, and the project team. It is never reported to management. We encourage unit managers to apply balanced measures of motivation (“leadership”) and recognize and respect the fact that there are always “different strokes for different folks”, and that this technique is something most, but not all, supervisors will be able to utilize in a constructive manner.