An analysis of the performance of push, pull and hybrid production systems in manufacturing supply chains

This thesis presents the analysis of dynamic behaviour arising from differentproduction planning systems in manufacturing supply chains. Models of push, pulland hybrid production planning systems are developed using the system dynamicsmethodology, so allowing valid comparisons across the models. Although theseproduction planning methods are well discussed in the literature in the light of theirperformance in different manufacturing systems, this research has a differentapproach. The models allow experimentation using a variety of hypothetical demandpatterns as input, so as to reflect the current increasing uncertainty in customerdemand. Results from the simulation runs are analysed on a comparative basis.Aspects such as the stock levels, back-order levels and cumulative costs areconsidered as performance metrics. In supply chains today, demand amplification(commonly called the bullwhip effect) remains one of the most difficult issues toresolve. The volatility of this wave can be observed from downstream customersplacing orders, through retailers, distributors and finally the end member of thesupply chain: manufacturers. This is the point where the research can offer somehelp. Findings from the analysis will provide insight as to how this wave transmitsthrough manufacturers' operations under distinct production planning methods,together with the costs incurred by such instability. In addition, dynamic optimizationof parameter values in the models (to minimise cumulative costs) is shown to be auseful methodological tool for managers in making decisions concerning stock andbacklog policies