System Dynamics modelling of closed loop supply chain systems for evaluating system improvement strategies

In the past, many companies were concerned primarily with managing activities along the traditional supply chain in order to optimise operational processes and thereby economic benefits, without considering new economic and environmental opportunities in relation to the reverse supply chain and the use of used or reclaimed products. In contrast, there is now increasing interest among companies in reverse logistics and closed loop supply chain (CLSC) and their economic benefits and environmental impacts. In particular, the concept of CLSC views the reverse flow (i.e. the reverse supply chain) of reclaimed goods as integral to the forward flow (i.e. the traditional supply chain) to the consumer. At the end of the useful life of products, a reverse supply process is activated in which unwanted materials and products are recovered from end users to recapture some of their value. Therefore, planning for the forward flow of goods must take into account the recovered products. Three main processes that need to be considered are: (1) collection and distribution planning; (2) inventory control; and (3) production planning. In this thesis, our focus is the study of remanufacturing activity, which is one of the main recovery methods applied to closed loop supply chains. Specifically, we investigate and evaluate strategies for effective management concerning inventory control and production planning of a remanufacturing system. In order to pursue such a research objective, we model a production and inventory system for remanufacturing using the System Dynamics (SD) simulation modelling approach. Our primary interest is in the remanufacturing and returns processes of such a system. 3 As part of the development of the SD models, we identify the main factors, their influence relationships and the business/operational policies that affect the dynamic behaviour of the system. The returns process is modelled using significant factors which define: (1) the average period of time for which a product stays with its customer before it is returned (residence time); (2) the incentives offered by companies for the recovery of the used products (service agreement with customer); and (3) the behaviour of customers in returning used products (customer behaviour). Interestingly, combining these factors in a process model addresses the issue regarding the uncertainty in quantity and timing of returns in the reverse supply chain.<br><br>To our knowledge, a returns process modelled with such factors and their influence relationships is not readily available in the literature. For the same system, the remanufacturing process is modelled using such key factors as: (1) integrated remanufacturing/production capacity, (2) lead times, (3) backorder and (4) inventory coverage. Several policies that affect the dynamic behaviour of the system are defined in the modelling process using such factors. These modelled policies are included in order to improve the efficiency of managing production/remanufacturing and inventory activities in the process. This thesis also contributes to the field through the analysis of several scenarios combining the aforementioned factors and utilising simulation in order to evaluate strategies aimed at the optimum performance of the system. The evaluation results reveal that efficiency in managing inventory can be improved by increasing the returns rate (quantity of returns), which in turn can be achieved by reducing the residence time and increasing company incentives for the recovery of used products. At the same time, the uncertainty around the returns rate is significantly diminished by increasing those incentives that encourage customers to return used products. Other findings indicate improved efficiency in the remanufacturing process with higher remanufacturing capacity if the quantity of remanufacturable returns and the remanufacturing lead time are increased and decreased, respectively. Moreover, increasing the production lead time affects system performance more than does an equivalent increase in the remanufacturing lead time. 4 Case studies are used in this thesis in order to support some of the research findings and to further validate the developed models of the production and inventory system for remanufacturing. The selection of companies employed as case studies was based on their engagement in remanufacturing and returns processes, which made them useful for our research. Specifically, data and information were collected through interviews with company management representatives of the Australian Mobile Telecommunications Association, Fuji Xerox Australia and CEVA Logistics. These three companies are significantly involved in operational and management activities linked to reverse logistics and remanufacturing processes. The knowledge gained about these companies’ activities, coupled with the data collected from the ‘real world’, were useful for the development of the models of returns and remanufacturing processes as well as for the assessment of the research findings.