RECOMMENDATIONS FOR PRINCIPLES IN THE DESIGN OF CONTROL CENTRES

A general philosophy for the design of control centres and the involvement of ergo­nomic design factors will now be considered. This section presents a discussion of the principles as a complement to the recommendations for the different types of ‘knobs and dials’ to be found in some other parts of this handbook.

Principles in the design of control centres can be discussed from three differ­ent perspectives:

1. The system design

2. Participative design and action research

3. Usage of handbook data

2.4.1 System Design

Obviously, control room operators should have a good knowledge of the process they are operating. They must also be well aware of the task they are expected to carry out. This is also the basis for the planning of the control room and its organisation.

It is assumed that a realistic human resource (HR) deployment and organisation plan has been produced on the basis of the job descriptions. But before deciding the tasks, and therefore the HR deployment and organisation plan, it is necessary to have made reasoned allocations of the functions that are to be automated and those to be performed by the operator. A systematic allocation of functions is a very useful tool to stimulate creative thinking. By using this process, new alternative designs can be innovated. This reasoning ties in with Singleton’s (1966) general systems design model. However, it is questionable whether in an overall plan it is possible to generalise about staffing levels and organisation within the process industries and, subsequently, to make design recommendations for the control room hardware (for example, display equipment, controls, working surfaces, workplace design). It is therefore essential to use this form of systematic methods of design as a starting point. It is equally important to include a more participative process of design. (This is discussed more fully below.)

During the course of the project on Computerisation in the Process Industries mentioned earlier, a number of parameters were determined. Determining param­eters made it possible to describe different processes in a relatively unified manner and to thereby find a relationship for the staffing and organisation requirements for the process.

The following variables may be used to describe the process:

1. Organisation within the company and in process operation

2. Complexity and degree of continuity

3. Quantitative aspects

4. Fault handling

5. Dynamic aspects

6. Job aids

7. Converging/diverging processes

8. Information and control devices

9. Environmental and layout aspects

Organisation (1) in the process industries is usually orientated towards (a) opti­misation of operation, or (b) maintenance and reliability. In the first case (a), the operators’ work in the control room is mainly theoretical. In the second case (b), there is usually a combination of different types of work. Here, operators work partly in the control room and partly on supervision and maintenance tasks in the plant itself.

Complexity and degree of continuity (2) imply variables whereby one can describe how difficult it is for the operators to understand and get an overview of the process, and how well integrated are the different parts of the process. In turn, these variables depend on the size of the process and the understandability of parts of the process or the process as a whole. That a process has a high degree of continuity is directly connected with the ‘difficulty of understanding’ of the process. A high degree of continuity implies a low degree of discrete functional parts (batches) in the process. In other words, as the process is a natural whole, the operator cannot divide it into any natural functional components. A high degree of continuity also gives less time and opportunity for the operator to consider and make corrections compared with a batch process. This is particularly true where the throughput is relatively fast or where it is dependent on several complex process stages.

Quantitative aspects (3) cover the amount of information processing/informa – tion supervision that the operator is expected to carry out. In most processes, this is probably a particularly critical parameter in relation to staffing levels. It includes, for example, how many variables are monitored from the control room, how many automatic control circuits are monitored, and how many automatic control circuits the operators handle manually. It also covers the number of alarms and the level of direct involvement in the process. As a rough rule-of-thumb, it may be said that there must be at least one operator per 100 control circuits.

The whole fault-handling system (4) is critical for the operators, both in terms of the number of operators and the types of operators needed in relation to job skills, training, and so forth. One also has to be aware of the fact that fault handling should not only deal with severe incidents and accidents. All errors, minor faults, and criti­cal incidents are important warning signals and should be followed up and evaluated as a natural part of the daily routines of process operations.

Dynamic aspects (5) of the system refer primarily to the kind of time-related response the system has to different actions taken by the operators, both manually and in the form of defining set values on regulators. Response times of control/com – puter systems are also of interest. Systems with high inertia and long response times make the process more difficult to control (see also the introduction to Chapter 3).

Another important factor in determining the level of staffing and type of per­sonnel needed is the type of job aids (6) available. This factor also relates directly to hardware design. To a certain extent, this factor is also directly dependent on the choice of automation level (type of allocation). For example, both the staffing situa­tion and the character of the job itself will be affected if operators rely on handwrit­ten log books or on computer memories with printers. The use of predictors and simulators is also of considerable importance. However, this form of job aid has been found to be very uncommon in process industries. Simulators in the control room (on-line or off-line) increase the possibility for operators to build up their control skill and also for optimising the running of the system. The last three variables have not been dealt with here as they are dealt with in other parts of this handbook.