C I D - 1 3 3 I S S N 1 4 0 3 - 0 7 2 1 D e p a r t m e n t o f N u m e r i c a l A n a l y s i s a n d C o m p u t e r S c i e n c e K T H
Work Environment and Computer Systems Development.
Jan Gulliksen and Bengt Sandblad
CID, CENTRE FOR USER ORIENTED IT DESIGN
Författare: Jan Gulliksen and Bengt Sandblad
Work Environment and Computer Systems Development.
Report number: CID-133
ISSN number: ISSN 1403-0721 (print) 1403-073X (Web/PDF)
Publication date: August 2001
E-mail of author: email@example.com
URL of author: http://cid.nada.kth.se
Reports can be ordered from:
CID, Centre for User Oriented IT Design
NADA, Deptartment of Numerical Analysis and Computer Science
KTH (Royal Institute of Technology)
SE-100 44 Stockhom, Sweden
Telephone: + 46 (0) 8 790 91 00
Fax: + 46 (0) 8 790 90 99
AND THE DEVELOPMENT OF INFORMATION SYSTEMS
J. Gulliksena,b and B. Sandblada
aDepartment of Human Computer Interaction, Information Technology, Uppsala University,
P.O. Box 337, SE-751 05 Uppsala, Sweden
bCentre for User Oriented IT-design (CID), Royal Institute of Technology, Stockholm, Sweden
Investigations indicate that different types of work environment problems in computer supported work are
constantly increasing. Most efforts to improve the work environment are focused on physical or psycho-
social aspects, since these are easier to measure and provide efficient solutions to. Mental workload and
cognitive problems are of a more complex nature, and are therefore more seldom studied and solved.
Solutions to work environment problems are usually applied to already constructed work situations
through health programs, education, reorganizations, etc. The problems are seldom prevented by methods
applied early in the systems development process, i.e. before the artifacts have been designed and
implemented. This, and the following papers of the session, will focus on the need to integrate different
interdisciplinary methods in the analysis, design and evaluation of computerized work situations, aimed at
preventing work environment problems.
1.1 Work environment problems in computer supported work
Despite the increasing interest in work environment problems, we know that problems with badly
designed an inefficient computer systems in working life are increasing. Statistics from Sweden show that
in 1999 65% of all Swedish employed used a computer tool in their everyday work. More then 30% used
computer systems during more then 50% of their working day. Recent investigations also show that up to
70% of all employed in office and case handling work, where computer systems are used during most of
the day, report daily problems related to stress, physical and mental load.
A class of work environment problems that becomes especially important to consider in intensively
computer-supported work are the cognitive work environment problems. With cognitive work environ-
ment problems we mean when properties of the work environment hinder the persons performing the
work to efficiently use their skills. These hindrances, i.e. the cognitive work environment problems, are
often associated with the design of the information system. If the system has an inappropriate function-
ality, a bad user interface or if the user does not have the right competencies such problems occur.
Most activities aimed at improving the work environment for professionals in intensively computer
supported work situations are adopted very late, i.e. when physical or psychosocial problems already have
been manifested and reported. We believe that it is both important and feasible to address potential work
environment problems already during the development process. We know, however, that organizations
already have difficulties in addressing aspects related to the usability of artifacts under development. If
we add the challenge to also include work environment aspects in the systems requirements, the task will
most often become impossible in practice.
1.2 Research goals
In our research, some modern software development models and tools have been studied and new
requirements concerning their properties have been specified, in order to improve them. Some of the most
important additional factors that must be treated by the software development tools are: modeling of
future work processes, integrated development of work organization and IT support systems, user
interface design that can reduce mental work load and control of the work processes, education and
training of future users in order to prepare them for the new technical systems and for the new work
2. INTEGRATION AND THE DEVELOPMENT PROCESS
2.1. Usability and work environment requirements
The major goal for every professional involved in user interface development is, presumably, to develop
systems that are usable. This should be especially important for user interface designers, as their efforts
have a most immediate effect on system usability. We use the term usability as defined in the ISO 9241
standards on “Software ergonomics for office work with visual display terminals (VDTs)” Part 11 –
Guidance on usability:
“Usability is the extent to which a product can be used by specified users to achieve specified goals with
effectiveness, efficiency, and satisfaction in a specified context of use”. [ISO, 1998]
Several researchers have put the blame for the lack of usability on the users themselves not requiring
sufficient usability in the process of ordering or purchasing the systems [Thimbleby, 1999]. Studies of
requirement specifications with respect to usability and work environment aspects show that these
requirements (however, in practice very few) were not particularly useful in the development process,
mainly because they were not concrete and related to a measurable entity. Directed efforts to help the
users specify measurable requirements on usability and the work environment resulted in very few
measurable requirements, and of these most of them related to efficiency.
2.2. Integrated methods
Integrated design refers to methods in which all aspects of usability evolve together [Gould, Boies &
Ukelson, 1997]. Our view on integrated design is a further elaboration of Leavitt’s  integration of
organization, work, competence and technology to involve all aspects that can influence the development
of a computerized work environment, especially work environment aspects that are often ignored.
We have been experimenting with integrated methods by applying interdisciplinary cooperation in real
development settings, e.g.:
• Integrated analysis – Experts in organizational development, knowledge production processes, work
environment and human computer interaction performs semi-structured “observation-interviews” in a
real work setting in parallel sessions. Immediately after the sessions the experts document and
prioritize findings according to their relevance and the degree of seriousness of the discovered
• Integrated design – Experts in systems development, usability, human computer interaction,
organizational development, knowledge production processes, and work environment participate in
cooperative design sessions. In these sessions groups of users design prototypes in parallel sessions
mediated by the experts. They later reconvene to present their different design solutions for further
elaboration. The experts present and report their observations back to the development team.
• Integrated evaluation – A usability inspection is performed in which we use the various multi-
disciplinary competencies in the evaluation process.
• All preliminary results point to the fact that integrated design requires the introduction of the specific
competencies into the actual development work. The complexity of all aspects relating to the health
and well-being of a worker using a computer artifact cannot be introduced without bringing in experts
from these different areas.
2.3. A user-centered development process
Most systems development work is today performed according to some more or less industrial model.
The purchasing organization is specifying their requirements on the system as far as their knowledge
allows. Here requirements related to usability are rare. The development organization is doing its best in
trying to interpret the requirements and transforming them into a functioning system as efficiently as
possible. In this process user involvement is normally avoided to a large extent in order to minimize
project risks, repeatedly modified requirements, and time consuming iterations. However, this often gives
a bad outcome, because active user involvement is a necessity for applying user centered design.
For an organization to adopt new procedures for their systems development it is essential that the
development model is initiated and controlled by the organization itself. To our knowledge, no one has
ever successfully deployed a user-centered development methodology towards an organization that is not
appreciative. We have earlier performed reengineering work within a large governmental in-house
development organization with successful results [Gulliksen & Göransson, 2001]. With our help they
defined their ways of looking upon and performing user-centered design and to initiate further
development of their user-centered design process by themselves.
The next step is to include such experiences in the used development methodology. Normally the standard
development models available do not fulfill necessary requirements with regard to being user-centered.
We have made some studies on the benefits of commercial system development methodologies such as
the Rational Unified Process (RUP) and found severe problems with the process from a user-centered
design perspective [Gulliksen, Göransson & Lif, 2001].
A conclusion is, that if we want to be able to consider work environment problems, development of
competencies, work organization etc. in the development process, we must make sure that additional
methodological steps for this are included in the development model used.
2.4. Usability designers to improve the user centered process
Usability champions [Gould et al. 1997] are one way to introduce HCI knowledge into the development
work from within. We introduced the concept of the usability designer as a specific competence to keep
the development process user centered [Göransson & Sandbäck, 1999]. We have seen examples where
HCI researchers have apprenticed like usability designers and we have also acted towards organizations
that have employed their own usability designer as a resource in the development team. The best results
can be reached, as far as we have seen, when the usability designer works as a mentor inside an
organization that consciously works towards a user centered design methodology.
Due to the complexity of the task, the size of the organizations and the limited authority of the usability
designer in the organizations, we have so far not been able to have the usability designer to be responsible
for the work environment aspects as well. Work environment professionals requires rather different skills
and methods, why it would not be realistic to combine all usability and work environment responsibilities
in one person.
There is, however, a risk that the organization can take the usability designer as a hostage, not giving the
role the essential mandate to perform and control the required activities, but blaming the role for potential
usability and work environment failures. In our case, this has been overcome by the connection the
usability designer has with the researchers, but this cannot be the solution under normal development
conditions. It is therefore essential that this role is given enough mandate in the development organiza-
tion, so that the rest of the team has to follow the advice and guidelines.
2.5 Development of user competencies
Normally users are offered training in the use of the new tools developed, i.e. the handling of the new
information systems. This is of course important, but in practice not satisfactory performed. It proves to
be more important to concentrate on developing skills and to train the users in how to perform the new
work procedures. This includes the handling of the tools, but more important also how to work efficiently
in the new work setting. It is our experience that too little attention is given to the design of the new work
procedures and what the differences are between the old procedures and the new. If the new work
procedures are performed according to old traditions and standards, this will result in inefficient use of the
new information systems, something that in its turn will result in increased mental workload and stress.
2.6 Other aspects to consider in the development process
There are numbers of other important aspects related to usability and the work environment that should be
considered during systems development. And appropriate methodological modules must be included in
the development process to address these issues. Some of these aspects will be further addressed in the
following papers, and we will here only underline one such aspect: the design of the new work
organization. Very often the structure of the new organization that will be the result of the development
process is not consciously designed, but it rather appears. The design of the organization must be
considered during the entire systems development process. It must be designed, visualized in the
prototypes, tested, evaluated and repeatedly refined according to the work activity requirements as they
develop and evolve during the system development process.
There is a need to increase the awareness and maturity when it comes to consider usability aspects and
particularly work environment issues already in the system development process. To be able to do this
efficiently the following activities are needed:
• Increase the awareness both among the developers and the development organization as well as
among the users and user organizations.
• Improve competencies for the user organizations to order systems based on requirements concerning
usability and work environment aspects.
• Complement traditional software development models with methods and tools to consider work
We believe that it is most efficient to put the efforts into the development process instead of trying to
solve usability problems in products and systems. In this way the usability and work environment
problems can be prevented instead of solved when they are already manifested. The following are some
of the observations that we have made in our applied research:
• Skills and knowledge development. Education programs for developers can be successful if they
repeatedly are directed towards all participants in the development teams. The responsibility for
usability and work environment aspects must be given to experts and not as additional tasks to
developers with other basic skills and responsibilities. Organizational mentors can support individual
developers and teams.
• Process improvement. Existing development processes do not give enough support for the
integration of the required competencies in the development work. One solution is to provide detailed
process guidelines for user-centered design and criteria for the selection of users in the development
• Good examples. Despite all existing methods and tools for improving the work environment and the
usability of the computer support, what really has an effect in practice is the good examples that we
can study and use as an inspiration. Unfortunately there are rather few good examples out in “real
life” development of information systems.
This research has been financed by the Swedish Council for Working Life, the Swedish National Tax
Board and the Swedish National Social Insurance Board.
1. International Organization of Standardization (1998). ISO 9241 Ergonomic requirements for office
work with visual display terminals (VDTs) Part 11 – Guidance on usability. (International standard).
2. H. Thimbleby.(2000) On discerning users. In J. Gulliksen, A. Lantz & I. Boivie (eds.) How to make
user centered design usable. Technical report TRITA-NA-D0006, CID-72. Available at:
3 . J.D. Gould, S.J. Boies, J. Ukelson (1997) How to Design Usable Systems. In Helander, M.,
Landauer, T.K., Prabhu, P. (eds.) Handbook of Human-Computer Interaction, 2nd ed. pp. 231-254,
Elsevier Science B.V.
4. H.J. Leavitt (1958) Managerial Psychology. University of Chicago Press, Ltd. London.
5 . J. Gulliksen & B. Göransson (2001) Reengineering the systems development process for user-
centered design. In Michitaka Hirose (ed.) Proceedings of INTERACT 2001, Tokyo, Japan. IOS
6. J. Gulliksen, B. Göransson & M. Lif (2001) A User centered Approach to Object-Oriented User
Interface Design. In M. van Harmelen (ed.) Designing Interactive Systems: Object Modeling and
User Interface Design. Addison Wesley Longman.
7. B. Göransson & T. Sandbäck (1999) Usability designers improve the user-centred design process. In
Proceedings for INTERACT’99, Edinburgh, UK.