Abstract
Reverse logistics plays an important role in environmental conscious manufacturing, including recycling, disassembly, reuse, and remanufacturing. Integrating a forward logistic system and a reverse logistic system to form a bi-directional logistic system hence becomes significant. Subsequently, the Delivery and Pickup Problems with Time Windows (DPPTWs) has drawn much attention.Three current DPPTW schemes derived from the three main logistic strategies were reviewed in this study. Among them, the Simultaneous Delivery and Pickup Problem with Time Windows (SDPPTW) lacked for an appropriate model and appropriated test problems; thus, it was investigated first in this study. Then, two more general schemes: the Flexible Delivery Pickup Problem with Time Windows (FDPPTW) and the Fuzzy Flexible Delivery Pickup Problem with Time Windows (FFDPPTW) were thoroughly studied. For these problems, the corresponding mathematical models were proposed and validated by Cplex. Since they are NP-hard, the solution procedures entitled Coevolutionary Algorithms (CEAs) were developed for solving the generated test problems. The computational results reveal the excellent effectiveness and efficiency of the developed CEAs. Moreover, a comparative study on crisp schemes shows that the FDPPTW is the most flexible, efficient, and economical scheme for certain environment. For the uncertain case, the FFDPPTW based on a credibility measure was proposed in the form of the Chance Constrained Programming Model to facilitate the decision support based on the decision maker’s preference.