Two types of heuristic capacity control policies a·re evaluated with a SLAM simulation model of a materials requirements planning production system. The control policy decisions are based solely on the size of the queue, as measured in standard hours of work, at each work center in the production system. Several classes of product mixes and product structures are investigated, as well as several levels of the control parameters of each control policy. The results indicate that each control policy gives rise to a unique population of weekly labor, work in process, and inventory level. Product structure is also identified as a major variable in Materials Requirements Planning systems performance. Sensitivity analysis of the cost functions for each policy indicate the conditions under which it will minimize the sum of labor costs, work in process holding costs, and inventory holding costs. The simulation model, MRPSIM, is included with a user's guide. / M.S.

A hybrid methodology was developed to fairly compare functional and cellular production environments with respect to the production of machined parts which constitute the indivisible components of some final products. The methodology provides a means of designing each production environment at the lowest possible cost and then comparing the two environments with respect to cost and non-cost performance measures. The results show that the long-held belief that the cellular manufacturing or group technology method of production may be superior to that of the traditional functional or job shop layout may not be correct. A detailed comparison using four problem sets with different job and machine mixes failed to indicate a clear case in which the cellular environment performed better than the functional. The methodology consists of two stages. Stage one has six hierarchical steps which systematically determine machine requirements and layout planning of each environment through mathematical modelling. External and internal operation constraints and inputs such as stochastic daily demand and operation times were considered. Stochastic programming was used in handling uncertain daily demand and operation times by specifying a desired minimum probability of meeting the demand for each job type in both environments. The MPSIII package was used in solving large mixed integer problems that resulted once nonlinear terms, due to the chance-constrained nature of the segments of the models, were linearized. Because of the large problem sizes, MPSIII input files had to be created using FORTRAN codes. In stage two, the SIMAN simulation language was used to determine the feasibility of stage one decisions and to obtain other system information. In simulation, some approximations were made to implement stage one decisions. For example, jobs received an average processing time in each operation class area rather than the exact operation time of the specific machine type to which the jobs were assigned in stage one. The effect of material handling distances and the use of limited number of work-in-process carriers were considered. Although the methodology was mainly developed for the comparison of the two production environments, it is readily usable for individual design of either production environment. In addition to the two main stages of development, this research also required the development of two other procedures: unitizing daily demands and the modifying the previously available job/cell grouping methods. / Ph. D.