Supervisory Control Task Research Articles

Alarm-initiated activities: an analysis of alarm handlingby operators using text-based alarm systems in supervisory control systems

Alarms constitute a significant aspect of the operator-system interface in human supervisory control tasks: they have the potential to display information that is of critical importance to overall system safety and integrity. This paper considers the trend toward the use of visual display terminals for the presentation of alarms, specifically in the use of scrolling text displays. It is suggested that this form of information display may not best serve all aspects of alarm handling. To illustrate why this is so, an analysis of human alarm handling activities is presented. The focus on text-based displays and annunciators allows the authors to concentrate on a number of problems associated with the alarm design using a specific medium. From this discussion, the alarm initiated activities model is developed. By considering the activities that people engage in when dealing with alarms, an insight is offered into the information requirements that can be used to inform the design of alarm systems.

Information Exchange and Team Coordination in Human Supervisory Control

This paper addresses issues of “socio-technical” demands for effective team performance in supervisory control tasks. Team members must exchange data and system information, as well as interpersonal meanings and contexts, in order to develop shared and synchronized models of system behavior. The author’s extensions of current human performance models, including SRK frameworks, incorporate multiple information exchange and task coordination performance issues required to support expert team performance and “optimal” information exchange in a dynamic task environment.

An intelligent human-machine system based on an ecological interface design concept

It seems both necessary and promising to develop an intelligent human-machine system, considering the objective of the human-machine system and the recent advance in cognitive engineering and artificial intelligence together with the ever-increasing importance of human factor issues in nuclear power plant operation and maintenance. It should support human operators in their knowledge-based behaviour and allow them to cope with unanticipated abnormal events, including recovery from erroneous human actions. A top-down design approach has been adopted based on cognitive work analysis, and (1) an ecological interface, (2) a cognitive model-based advisor and (3) a robust automatic sequence controller have been established. These functions have been integrated into an experimental control room. A validation test was carried out by the participation of experienced operators and engineers. The results showed the usefulness of this system in supporting the operator's supervisory plant control tasks.

An Intelligent Man-Machine System for Future Nuclear Power Plants

The objective of the development of an intelligent man-machine system for future nuclear power plants is enhancement of operational reliability by applying recent advances in cognitive science, artificial intelligence, and computer technologies. To realize this objective, the intelligent man-machine system, aiming to support a knowledge-based decision making process in an operator’s supervisory plant control tasks, consists of three main functions, i.e., a cognitive model-based advisor, a robust automatic sequence controller, and an ecological interface. These three functions have been integrated into a console-type nuclear power plant monitoring and control system as a validation test bed. The validation tests in which experienced operator crews participated were carried out in 1991 and 1992. The test results show the usefulness of the support functions and the validity of the system design approach.

Dual-arm supervisory and shared control task description and execution

A unified control method for description and execution of cooperative dual-arm telerobotic supervisory and shared control tasks with multiple sensory inputs is presented. A general task execution primitive provides the capability. The desired behavior of execution is described by a large input parameter set to the primitive. The method provides a wide range of capabilities within the projected constraints for near-term space telerobotics systems such as limited computational power and safety. Move-squeeze decomposition is utilized to decompose forces sensed at the wrists of the two manipulators into forces in the move subspace, which cause system motion, and forces in the squeeze subspaces, which cause internal forces. The move and squeeze forces are then separately controlled. Each sensor control source is provided with its own task space for control before merging the inputs in a common move space frame. Bounded behavior control is used to ensure execution within safety bounds as specified a priori to task execution. Several space servicing tasks utilizing the cooperative dual-arm control capability are described and experimental results from the tasks are given. The supervisory and shared control tasks include capture of a rotating satellite, orbital replacement unit changeout, fluid coupler seating and locking, and contour following.

Supervisory control in a dynamic and uncertain environment: laboratory task and crew performance

A research program using a microworld methodology for investigating human performance in a dynamic uncertain environment is described. One- and two-person crews performed a supervisory control task that involved piloting a low-fidelity scout helicopter and supervising four additional helicopters. The simulated environment contained cargo and enemy craft that had to be discovered by searching a 100 mi/sup 2/ forested region. The task required planning, intervention, and several types of resource management. An expert one-person crew performed comparably to novice two-person crews, and both of these were superior to novice one-person crews. More detailed analyses of performance revealed several types of limited adaptivity in the behavior of the novice crews.< <ETX xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">&gt;</ETX>

Can speech be used for alarm displays in ‘process control’ type tasks?

Abstract There has been much research into the feasibility of speech in aircraft cockpits, but little in human supervisory control tasks. Speech displays can provide a number of benefits over conventional, visual displays, particularly as a means of providing alarm information. We discuss the term ‘alarm’, and suggest that different alarm situations will have different information requirements. Thus, a single type of alarm display may not be suitable for the complete range of situations encountered in the control room. We investigated the use of speech for different ‘alarm-initiated actions’: recording, urgency rating, location identification, and action specification. These tasks varied in terms of difficulty, and this affected performance. We also varied the quality of speech, comparing synthesized with human speech. While speech quality affected performance on the recording task, we found that task difficulty interacted with speech quality on the other tasks. This means that definable ‘trade-offs’ exis...

Human-Computer Cooperative Problem Solving in Satellite Ground Control

In the supervisory control of highly automated dynamic systems, the human operator is responsible for monitoring system state and compensating for system failures. Increasingly, the human operator interacts with one or more knowledge-based (e.g., expert) systems that assume responsibility for a portion of the supervisory control task. Thus, supervisory control is shared by both human and machine agent(s). To date, however, there is little research addressing the interaction of human and knowledge-based machine agents in supervisory control. This paper describes research that explores issues in the design of the cooperative human-machine interaction in the control of a complex dynamic system. The paper presents a set of prescriptive principles that define the human-computer interaction in supervisory control systems. In addition, the paper summarizes a case study of a NASA satellite ground control system in which human operators work with an expert system. The prescriptive principles and case study results form the basis of an architecture for cooperative problem solving for real-time control of dynamic systems.

Operator Models for Supervisory Control Systems

This paper presents a conceptual discussion of four human operator models that are potentially useful for supervisory control applications: the operator function model (Mitchell, 1987), the problem behavior graph (Newell and Simon, 1972), the decision ladder (Rasmussen, 1986), and goal-means network (Woods and Hollnagel, 1987). These models are characterized along the dimensions proposed by Jones and Mitchell (1987) and are further examined in-depth with the use of verbal protocols collected concurrently with the performance of a supervisory control task.

Evaluating the Accuracy of a Large-Vocabulary Speech Recognition System

This paper presents a conceptual discussion of four human operator models that are potentially useful for supervisory control applications: the operator function model (Mitchell, 1987), the problem behavior graph (Newell and Simon, 1972), the decision ladder (Rasmussen, 1986), and goal-means network (Woods and Hollnagel, 1987). These models are characterized along the dimensions proposed by Jones and Mitchell (1987) and are further examined in-depth with the use of verbal protocols collected concurrently with the performance of a supervisory control task.

Intelligent warning systems for instrument landings

The requirements of maintaining supervisory control of a complex system may sometimes exceed the capability of the human operator for a variety of reasons ranging from temporary distraction from the main task to excessive mental workload. The ergonomic aspect of complex supervisory control can be improved by augmenting the interface with expert systems. This paper outlines the development of an intelligent warning system as one component of an augmented supervisory control system for complex aircraft. The system is designed to alert the pilot to potentially hazardous conditions during instrument landing approaches. The process of task analysis that precedes the implementation of the system is outlined, drawing on the experience of a former military pilot. The results and methods shown here can be generalized to other components of the pilot's task and, more generally, to other complex supervisory control tasks.

Part-Task vs. Whole-Task Training on a Supervisory Control Task

The primary aim of training is to improve performance. Part-task training may be the more economical method, because full mission training simulators often cost more than the vehicles they simulate. However, the skills learned may not transfer effectively to performance of the complete task. This study investigated the effectiveness of Part-task training on the psychomotor portion of a supervisory control simulation. Twelve subjects were divided into Part-task and Whole-task groups and told to perform the task as quickly as possible. Part-task training was provided with the cursor-control device (a magnetic pen and pad), prior to transition to the Whole-task. Some distinct advantages of the Part-task training were: (1) The Part-task group learned the task faster; (2) The Part-task group's scores and task times continued to improve, while the Whole-task group's did not; and (3) A significant increase in speed of response for the Part-task group and almost no improvement in speed for the Whole-task group.

Effects of Functionally or Topographically Presented Process Schemes on Operator Performance

Two presentation modes of process schemes were compared in a supervisory control task. Operators Monitored a submarine's trim system and had to identify process states and execute control orders. It was expected that the presentation of a functional arrangement would result in shorter identification times and fewer control errors than would a topographical arrangement. Results showed no differences between the two presentation modes with regard to the time required to identify a particular process state. The time needed to perform control settings was shortest in the functional presentation mode, but this effect decreased with practice. In both presentation modes, experienced subjects performed better than inexperienced subjects. The results were in line with expectations, except for a significant error increase with practice in the functional presentation mode. It is suggested that this effect is due to subjects relying increasingly on memory rather than on the presented information.

A Discrete Control Model of Operator Function: A Methodology for Information Display Design

Recent advances in computer technology and the changing rule of the human in complex systems require changes in design strategies for information displays. The use of discrete control models to represent the human operator's cognitive and decisionmaking activities is described. The analytic procedures required to build a discrete control model show promise as a basis of a design methodology for the definition of an information display system for supervisory control tasks. The discrete control modeling procedures and their application for a simulated system is demonstrated.

Modeling Supervisory Pilot Behavior with General Systems Theory Formalisms

In a simulation study, conducted to determine information flow at a pilotcockpit interface during an automated landing approach, a mathematical model of pilot's supervisory control tasks was developed. For modeling those tasks and their network combination, production systems were used and described with general systems theory formalisms. For that the pilot was considered as a sequential input/output system and described with input, output, internal, and memory variables. While performing a task, only subsets of all variables and of their value sets are required. For describing the pilot's selection behavior, selection functions were introduced which specify task relevant variables and their values. From task relevant variables, behavior relevant variables are derived. Then the task performance behavior of the pilot can be described with binary relations among all value sets of behavior relevant variables which are task specific and invariant during the duration of a task. Each relation element represents an ordered pair of coordinates of which the first describes the left situation side of a production rule and the second its right action side. The resulting model guides the required task analysis and supports the development of a simulation program.

Application of Preference Tree Methodology to Choice Behavior in a AAA Supervisory Control Task

An attempt was made to extend a model of preference, known as elimination-by-aspects (Tversky, 1972), to situations involving active search for sequentially available information of varying value. Subjects were instructed to choose the most appropriate one of three possible operating modes in a task simulating supervisory decisions in a AAA system. The data failed to support the EBA model. Instead of searching aspects associated with each alternative in a non-compensatory fashion, as predicted by the model, subjects searched the information set in a compensatory manner. Failure of the model is discussed in terms of three points (a) subject control of information search, (b) sequential information availability, and (c) variability in the value of information across conditions. Consideration of these three points leads to the conclusion that the usefulness of the EBA approach is limited to situations in which the subject has all relevant information available to him before consideration of alternatives.

Copenhagen Central Mail Sorting Office Computer Control

A description is given of the structure and organization of the computer control in the Copenhagen Central Mail Sorting Office. This comprises control and monitoring of extensive transportation and sorting equipment by means of a number of computers operating at several levels and in accordance with a hierarchical system concept.The process data communication is mainly taken care of by ring-buslines equipped with a total of about 1,000 control terminals which are used for both input and output. A brief outline is given of the most important control tasks (automatic control, operator control and reporting), together with a more detailed account of the supervisory control tasks performed by the central computer.