Adaptive Production Planning and Scheduling for the Make-to-order DNA Manufacturing System

Abstract

This thesis develops an adaptive production planning and scheduling system for the make-to-order plasmid (DNA) manufacturing system. The system, which has stochastic nature and random demand, was represented by a mathematical programming model first. Then in order to solve it, discrete-event simulation models were developed to generate a feasible schedule that maximizes the production throughput in the planning horizon in a mix-product type environment. A special heuristic order selecting and splitting procedure was designed to aid the production planning and scheduling process. Experiments were conducted to evaluate the algorithm and results are compared with those obtained by using four classic dispatching rules, such as first come first served (FCFS) and shortest processing time (SPT). To take advantage of simulation results, a rule-based expert system was created with pre-defined scheduling rules. Rules regarding production planning and scheduling can be used by human schedulers easily and the system is very flexible in further extension.