Lot size scheduling problem with two level setup cost/time structure

We address the problem of scheduling production on a single capacitated processor when there are multiple product families with two level setup cost/time structure. The setup of the Family Lot Scheduling Problem (FLSP) has nonzero costs and times for both major setups between families and minor setups between products within a family. Families are defined by collecting products according to setup characteristics of the manufacturing process and products. Efficiency and savings can be achieved by coordinating the production on a family-by-family basis in order to reduce the number of major setup costs and times incurred.; In this dissertation, family lot size scheduling as an application of the Group Technology (GT) concept to production planning and scheduling has been proposed in order to get benefits from GT in the area of production and operations management.; To understand fundamental concepts in lot size scheduling, the literature of the Economic Lot Scheduling Problem (ELSP) is reviewed and solution methods identified and classified. We give a complete formal description of the generalized basic period algorithm for solving the ELSP. A heuristic method for solving the FLSP in a deterministic and stationary demand pattern is then developed and tested. First, a family schedule is created using aggregated data, and then schedules for products within families are obtained.; We next treat the FLSP when demand varies over time in a deterministic manner. Depending upon assumptions about the sequence dependent nature of family setup, FLSP is classified and formulated as a mixed-integer programming model. For a particular setup structure, a heuristic algorithm is developed and its performance evaluated using a set of test problems.; Finally, the FLSP under stochastic demand is considered. The limited literature of the stochastic ELSP is reviewed and extension to the stochastic FLSP discussed.