New Solution Approaches for Scheduling Problems in Production and Logistics

The current cumulative PhD thesis consists of six papers published in/submitted to scientific journals. The focus of the thesis is to develop new solution approaches for scheduling problems encountering in manufacturing as well as in logistics. The thesis is divided into two parts: “ma-chine scheduling in production” and “scheduling problems in logistics” each of them consisting three papers. To have most comprehensive overview of the topic of machine scheduling, the first part of the thesis starts with two systematic review papers, which were conducted on tertiary level (i.e., re-viewing literature reviews). Both of these papers analyze a sample of around 130 literature re-views on machine scheduling problems. The first paper use a subjective quantitative approach to evaluate the sample, while the second papers uses content analysis which is an objective quanti-tative approach to extract meaningful information from massive data. Based on the analysis, main attributes of scheduling problems in production are identified and are classified into sever-al categories. Although the focus of both these papers are set to review scheduling problems in manufacturing, the results are not restricted to machine scheduling problem and the results can be extended to the second part of the thesis. General drawbacks of literature reviews are identi-fied and several suggestions for future researches are also provided in both papers. The third paper in the first part of the thesis presents the results of 105 new heuristic algorithms developed to minimize total flow time of a set of jobs in a flowshop manufacturing environ-ment. The computational experiments confirm that the best heuristic proposed in this paper im-proves the average error of best existing algorithm by around 25 percent. The first paper in second part is focused on minimizing number of electric tow-trains responsi-ble to deliver spare parts from warehouse to the production lines. Together with minimizing number of these electric vehicles the paper is also focused to maximize the work load balance among the drivers of the vehicles. For this problem, after analyzing the complexity of the prob-lem, an opening heuristic, a mixed integer linear programing (MILP) model and a taboo-search neighborhood search approach are proposed. Several managerial insights, such as the effect of battery capacity on the number of required vehicles, are also discussed. The second paper of the second part addresses the problem of preparing unit loaded devices (ULDs) at air cargos to be loaded latter on in planes. The objective of this problem is to mini-mize number of workers required in a way that all existing flight departure times are met and number of available places for building ULDs is not violated. For this problem, first, a MILP model is proposed and then it is boosted with a couple of heuristics which enabled the model to find near optimum solutions in a matter of 10 seconds. The paper also investigates the inherent tradeoff between labor and space utilization as well as the uncertainty about the volume of cargo to be processed. The last paper of the second part proposes an integrated model to improve both ergonomic and economic performance of manual order picking process by rotating pallets in the warehouse. For the problem under consideration in this paper, we first present and MILP model and then pro-pose a neighborhood search based on simulated annealing. The results of numerical experiment indicate that selectively rotating pallets may reduce both order picking time as well as the load on order picker, which leads to a quicker and less risky order picking process