A decision-making tool to identify routings for an efficient utilization of machining resources: the decision makers’ perspective

Paper in proceeding, 2020

In the aerospace industry, efficient management of machining capacity is crucial to meet the required service levels to customers (which includes, measures of quality and lead-times ) and keeps the tied-up working capital in check. The proposed decision-making tool, described in this paper, aims to combine information and knowledge of manufacturing and logistics experts in a company to improve flow of materials through the factory. The material flow situation is different for a large aerospace tier-1 supplier as opposed to flow-based manufacturing company; when there is no pandemic or natural calamity, having relatively stable demand due to long-term contract is common, but there exists short-term demand variability. There is a complex flow of products at GKN Aerospace, as the products share machining resources, thus, resulting in uneven loads at machines and sometimes excess loading at certain machines. This along with short-term demand variability results in long queues in-front of machines which contributes with the biggest share of the total lead-time. Thus, long waiting times at one/many machine/s is common and may lead to bottlenecks in many places in the production pipeline. So, there is potential benefit in having rerouting-flexibility for products which can help in reducing queuing. However, qualifying a product for a new machine is time-consuming activity, and thus, should be done few years in advance.

We propose a mathematical model aimed at improving some of these deficiencies of commonly used methods by facilitating balanced resource loading levels, i.e. to provide more degrees of freedom to the planner to absorb demand variations. The output provided by the model includes production routings in each time period (quarter) for the next 4–5 years; new qualifications to be done by technical staff for allocation of part types/products to machines which are not yet qualified/used for a given product. We keep the resource loading levels that are above a given threshold as low as possible and reduce the time/money spent for qualifying new allocations.