How to define the tool kit for the corrective maintenance service? a tool kit definition model under the service performance criterion

Currently, the rule of defining tool kits is varied and more engineer's aspects oriented. However, the decision of the tool kit's definition is a trade-off problem between the cost and the service performance. This project is designed to develop a model that can integrate the engineer's preferences and logistics perspectives and propose the solution for a good definition of tool kit. In the high level analysis, we investigate the potential DTWT saving by having tool kits under different uncertainty probabilities. The uncertainty probability implies the probability that engineers find unexpected problems during repair and consequently need extra tools. From the result, under the high uncertainty probability, having tool kits obviously results to a high DTWT saving; moreover, high investment in tool kits leads to a better DTWT saving in terms of better service performance. However, the results of the analysis are valued when the system has well defined tool kits, and the investment is worth when the DTWT saving is needed for the system. Therefore, a decision model is developed regarding the following critical perspectives. • From the engineer's preferences: (1) characteristics of the repair action, (2) packing, (3) uncertainty, (4) time related to uncertainty, and (5) convenience. • From the logistics perspectives: (1) cost including operational cost and investment cost, (2) service level, and (3) demand. This model aims to decide a tool kit for a FC/BB, and the decision is to choose tools from a candidate tool list to compose a tool kit. This candidate tool list consists of the needed sets of tools corresponding to the service requirements for this specific FC/BB. Regarding the above critical perspectives, the model integrates the following parameters. • Operational Cost The operational cost includes the shipment cost of tools, the cleaning cost of tools, and the handling cost of tools. • Investment Cost The investment cost includes the purchasing cost of tools and the package cost of the kit. • Service Performance The service performance contains the shipment time of tools. • Size and Weight The tool's size and weight are included. To incorporate these parameters, the other variables are needed to generate the overall cost and the overall service performance: (1) probability of ordering each set of tools; (2) uncertainty probability; (3) number of installed machines; (4) failure rate of the specified FC/BB; (5) size of the service network. The model is formulated as searching the tool kit definition with the minimum total cost under the criteria of the service performance and the physical constraints of tool kit. Since the model is complex, the heuristic method has been used to find solutions and the model is for the evaluations. The model is applied in two real-life examples and their results are presented below. 1. The tool kit in NPI phase • Result The proposed definition includes two tools {2, 6} compared with 13 tools in the tool kit defined by engineers. This proposed tool kit can save 119,265.76 € which is 34% cost reduced from the tool kit defined by engineers. The main cost saving is from the operational cost. • Scenario tests In this example, several scenario tests are designed for the following parameters. i. Target of FCDTWT By setting different targets, we obtained different definitions. With the stricter target, more tools are included and the higher cost is obtained. By adjusting the target, decision makers can decide how much budget they can spend on the tool kit for the specific FC/BB, and what the expected performance of this tool kit would be. ii. Probability of the first order and handling cost Through the scenario tests, the solution from the model is robust to against the impacts from these two parameters. iii. Uncertainty probability and shipment time Through the scenario tests, the solution would be affected by these two factors, especially the shipment time. Although the uncertainty probability is hard to predict at this moment, the engineers can put a high uncertainty e.g. 50% to define the tool kit. About the shipment time, the more precise values of the input data yield a better quality of definitions. 2. The tool kit in IBL phase • Result Based on the collected data from the survey and logistics department, the model suggested that no tool kit is needed. Comparing with 18 tools in the tool kit defined by engineers, no tool kit can save 205,821.51 (€) which is 30.26% cost reduced from the tool kit defined by engineers. The main cost saving is from the operational cost. From the numerical results in two tool kits, both of the proposed definitions can save more than 30% of cost, mainly in the operational cost. Because the significant cost saving from the unnecessary tool's shipment can be done by having smaller tool kits in the system. This project is concluded by applying some assumptions such as tool kits should be located in regular local warehouses, and the inventory policy is out of scope. Besides, we relaxed the constraints of weight and size while searching the solutions. In the future research, the model can be extended by breaking these assumptions, and applying the physical constraints to search the solutions. To sum up, this project provides the model for the tool kit definition and the heuristics to find the proposed solution. This model and the heuristics are programmed in Excel by VBA, and it can be easily executed by the decision makers. The applications on two real-life examples are done and analyzed by several scenario tests. In these two examples, the results indicate that the current tool kits can be improved by having fewer tools inside and still can satisfy the service performance. Moreover, the proposed tool kits in both examples can save more than 30% of the total cost from the current tool kits. It implies the significant improvement could be done by this tool kit definition model.