An Expert System Design for Alternating Job Shift Scheduling*

In around-the-clock services such as hospital patient care, workers must be assigned fairly to different job types (called shifts), which require various number of workers for a given period of time. This paper proposes a design for an expert system to do this scheduling, under the condition of assignment-inhibiting irregular events such as sick leaves. In this design, the hierarchical problem-solving approach of a human expert is incorporated as a heuristic to reduce the number of searches required to find an optimum solution. That is, a rough solution is derived under the dominant constraints using an algorithmic assignment method called ARA followed by leveling of the resulting unbalanced work load by a trial-and-error replacement method of workers called MMR