Operator Workload Knowledge-Based Expert System Tool (OWLKNEST) Demonstration

Abstract

Objectives The Operator Workload Knowledge-based Expert System Tool (OWLKNEST) is a tool that provides guidance in selecting the most appropriate technique(s) for estimating or predicting Operator Workload (OWL). This demonstration will provide hands-on usage for interested parties in utilizing OWLKNEST to determine the most appropriate OWL technique for their particular situation, interpreting the resulting outputs, and performing sensitivity analysis to assess the impact of changing responses. Description of subject matter A variety of OWL estimation techniques are available, but information about these techniques and their appropriateness for particular situations may be difficult to assimilate. Furthermore, information on practical issues concerning the applicability of a particular tool is often unavailable—e.g., time, cost, personnel issues, etc. The selection of the appropriate technique(s) for assessing OWL is complex and beyond the scope of many practitioners. Therefore, an expert system approach was selected to provide a tool which can serve as a clearinghouse of knowledge as well as providing recommendations for workload estimation techniques. OWLKNEST requires an IBM PC type microcomputer equipped with a minimum of 640 Kb memory and two floppy diskette drives of at least 360Kb. Exsys Professional, a rule-based, backward-chaining system, is the expert system shell used for OWLKNEST. OWLKNEST incorporates an hierarchical logic structure that quickly focuses on the most applicable technique(s) and minimizes the number of questions posed by the system. It is based upon a taxonomy which divides OWL techniques into analytical and empirical techniques. OWLKNEST utilizes a question-and-answer dialogue to facilitates use by inexperienced analysts and is supplemented by embedded help features. The inputs are fed into the expert system, which applies rules and knowledge based on user information. The result is a suggested list of appropriate techniques. The user can obtain brief descriptions of the recommended technique(s) including references and descriptions of prior applications. OWLKNEST assumes that users have at least a fundamental knowledge of OWL concepts but can be novice computer and expert system users. The development of OWLKNEST was sponsored by the Army Research Institute and free copies of the runtime version of OWLKNEST are available to interested parties. Additional details about OWLKNEST are available in Hill and Harris, 1989 and Harris et al, 1989. More complete information about utilizing OWLKNEST can be obtained in the user's guide—Handbook for Operation of the OWLKNEST Tool (HOOT) (Harris et al., 1989). Importance The increasing criticality of performing OWL assessments throughout the system development cycle has Id to the recognition that a comprehensive, easy-to-use tool to recommend the appropriate OWL techniques is needed. OWLKNEST provides this capability and supplies information on the appropriate OWL techniques based on individual needs and resources. Possible Applications OWLKNEST will be useful in military application as well as other application areas where prediction and evaluation of operator workload is of interest. Similarly, the workload techniques are not application specific, but can be used in a variety of domains. OWLKNEST can provide the information and guidance of an appropriate selection of workload techniques for a broad range of applications.