To Develop a Knowledge-Based System for Parallel Machine Operations Scheduling

As the industrialized world develops, more and more resources are becoming critical.
Machines, manpower and facilities are now commonly thought of as critical resources
in production and service activities. Scheduling these resources leads to increased
efficiency, utilization, and ultimately, profitability. Current operation scheduling
methods require complex mathematically modeling techniques that demand the
substantial and extensive knowledge. Otherwise, the simple methods may not provide
good results. The project is intended to engage in the issue of parallel machine
operation scheduling from the knowledge based system perspective. The deal between
both fields will emerge a new development of formulation and integration in artificial
intelligence at area of industrial scheduling. Eventhough there are diversity techniques
in manufacturing industry, the scope of the study is only limited to identical parallel
operation scheduling due to time constraint towards the completion of the project. The
goal of this project is to produce a working model of knowledge system for parallel
machine operation scheduling. The execution of the project will be conducted in two
semesters. For the First Semester, the data gathering and carrying out the associated
case studies were explicitly nurtured as to aid better understanding of the project. The
case studies performed were: (1) Parallel Processing - Jobs of Equal Weight, (2)
Parallel Processing - Weighted Jobs, and (3) Parallel Processing - Jobs with Due Dates.
In the Second Semester, the knowledge system was effectively developed. The
development of the knowledge system was done in the expert interface which consists
of six parts. The first step is theory familiarization, followed by user interface
development, the inference engine development, dry run or testing and verification of
the system. When all the steps are taken, the knowledge system can be considered as
complete.