Late registration: Bjarne.Foss@itk.ntnu.no or meet up at the registration desk in the morning.
The participants will be divided into groups of 4 to 8 members. Each group will choose 1-2 problem areas from the following problem set:
Problem area 1: Learning support systems
Problem area 2: Modeling as a learning process
Problem area 3: Integration of design and operations
Problem area 4: Development and commisioning of advanced support systems
for plant operations.
The aim of the group work will be in-depth discussions which hopefully will reflect adiversity in background and experience and, hence, promote some learning on the complex issues pinponted above. Further, each group will be asked to present their findings in the final plenary session.
Problem area 1: LEARNING SUPPORT SYSTEMS
In today's global and turbulent business environment, learning has attracted attention as a means for companies to stay competitive. Arie De Geus, former head of planning in Royal Dutch/Shell put it this way: ?The ability to learn faster than your competitors may be the only sustainable competitive advantage.? In this group, we will discuss the role of process information and control systems in a learning perspective.
The role of human beings in the operation of modern process plants, can be seen from two opposite points of view:
1. Operators are present in the plant to cope with process deviations and alarm conditions, and otherwise leave the control of the process to the automation system. This is the MECHANISTIC view.
2. Operators represent a flexible capacity to accumulate competence and knowledge, to be able to meet changing conditions and future challenges, that may be introduced by the plant external environment, i.e. customers and authorities. This is the LEARNING view.
We will discuss how these two opposite views on human beings in plant operation influence the information systems used for running a process plant. Traditionally, process control and information systems (PCISs) have the mechanistic view as design basis. Take for instance, the use of 'black-box' technologies; When an operator cannot 'see' and understand how things are operating, he will not learn. Expert systems are a typical example of such 'black-box' solutions. Another example is process control systems in automated plants; The operators will often encounter passive working conditions. This does not support learning.
A learning support system (LSS), on the other hand, origin from the learning point of view. The aim of such a system is to support, not replace, the operator. The LSS is a kind of process control and information system that has not yet been developed.
Suggested questions for discussion
1. Are learning support systems an interesting and relevant issue?
2. In what ways can a process control and information system contribute to
learning?
3. What does a LSS look like; should it be different from a traditional PCIS?
4. Eventually, what are the limitations (technical, economical, etc.) when
developing a LSS?
Problem area 2: MODELING AS A LEARNING PROCESS
Flexibility, adaptability and knowledge are expected to be some of the main competitive factors in the future. This again raises a continuous need for learning and development, -- for the individuals and the organizations. Methods for knowledge development depend on limiting conditions such as culture, what kind of enterprise we are dealing with, history, resources etc.
This problem area raises the question on whether modeling is a suitable method for learning and development. Modeling is here the process of creating a representation or a mathematical model of the topic studied, e.g. the production process or work procedures. The modeling process is in this discussion based on two-way communication. It is a process where both experts and end users work together for a common goal. It is not the experts job to gather the information they need from the users to perform the modeling. This means that the modeling process is not only the creation of a model/representation, but also a process of gathering and sharing of information.
We would in this session like to discuss the modeling process as a tool for learning and development, and improvement. To improve something, we have to make some changes. Changes is easier to implement if they are accepted and understood by whom it may concern. This is the idea behind using modeling as a tool for learning and development. To the modeling process we may achieve acceptance and common understanding of the enterprise of the issue we want to change.
Discussion subject:
Is modeling one of the areas of core competence in the process
industry today? Is it an important strategic activity for the
enterprise?
Suggested questions:
1. Is learning and development an important issue for the process
industry? Why, or why not?
2. Can modeling be used as a tool for learning and development?
3. In what kind of situations can modeling successfully be used?
(e.g. improvement of production process, modeling of work processes,
optimization)
4. How can a modeling process be run to enhance learning?
Who should participate?
What conditions must be fulfilled?
Does a permanent improvement require much more, -- like expert programming and implementation?
Why should the operators (the "users") participate in the modeling process?
What can possible advantages of a modeling process with both experts and operators/users be?
Problem area 3: INTEGRATION OF DESIGN AND OPERATIONS
Reduction of maintenance costs and increased operations effectiveness may be arguments for integrating perspectives of operations and design. This is of particular interest to the offshore industry today where the environments of operations and design traditionally have been firmly separated from each other. Operations and maintenance experience have tended to be kept within the operating companies, without any systematic gathering of operational data and knowledge. Design has been performed onshore by engineers that rarely have any operations and maintenance experience themselves, and who seldom get feedback from operations on what they design. There has not been much communication between people of operations and design.
While people in operations tend to have little formal education but much practical knowledge, engineers in design have an academic background focused on theory. This difference between practical and theoretical knowledge is one of the major barriers that people in operations and design feel and which complicates the dialogue between the two. Day to day operations require a certain amount of practical considerations - a kind of knowledge that may be difficult for engineers in design to understand. Without any operations and maintenance experience, it is difficult to value the importance of the practical reflections of operations and maintenance work. Availability and easy access to equipment are factors that are of major importance to operations, something which has been difficult to take account for in design. Practical operational knowledge is not necessarily respected and valued in design and through the engineering education, neither are people awarded for thinking in terms of operations in design. However, I would argue that it is when practical and theoretical knowledge of operations and design do meet that innovations are likely to occur.
Interesting questions to discuss may be:
What are the difficulties in terms of taking operations experience into
design?
What is operations in relation to design, and how does operations think
differently from design?
What are the barriers towards taking account of operations experience in
design?
How to bridge practical and theoretical knowledge in operations and design?
How to create fruitful meetings between operations and design?
When are innovations likely to occur?
Problem area 4: DESIGN AND IMPLEMENTATION OF ADVANCED SUPPORT
SYSTEMS FOR PLANT OPERATIONS
Today the operator in a process plant is faced with an increasing amount of advanced technology like experts systems, fault detection systems and other decision support systems. But also the control system itself incorporate more advanced functions like model based (predictive) controllers and alarmfiltering. Due to the increased level of automation, the operators are also having less manual interaction with the process. These advanced systems are often designed and implemented by engineers and other experts. It is observed, though, that after a while a fairly large number of these implementations tend to be inactive. The operators have either switched them off (or to manual), or are just not using them.
The main questions are then:
1. Why do an advanced system tend not to be used by the operators after a while
(i.e. the installation has failed)?
2. What can be done to increase the number of succesfull implementations?
(succesfull also one year after the installation)
Suggested issues to discuss are:
- participation and/or cooperation between engineers and operators in the
design phase
- level of education for the operators
- operator training
- maintenance of the system
- the operators role in the improvement or development of the next generation
of a support/control system, i.e. the life-cycle of the project
(design, implementation, operation, design, ...)
Group work guidelines
The general idea behind the group work is to create an option for discussing issues with colleagues. Differences in education, position, experiences and culture is what might make the group-work a creative and learning= experience.
We have suggested a number of topics. The idea is that the members start by discussing the field of interest aiming at making sense of the focus. It seems like it is important that everyone participate in this initial phase of the group work in order to give a voice to the different experiences that are present. You are not expected to come to any kind of conclusion in this phase, but just to share your intellectual position.
In the second part of the group-work we have suggested a number of questions to discuss. These are only "suggestions" from our point of view. The group has of course full freedom to make its own decision on how to use the time. The only constraint we put on each group is to be back for a plenary discussion at 1530.
The report back to the plenary discussion should not be a summary of the groups work, but we encourage each group to present to the plenary for discussion what they consider to be the most important message given the discussion in the group. Our aim is that the report back to the plenary what you really think is important.
Have an exiting and fruitful intellectual discourse in the groups!