University of Twente Student Theses




[img] PDF
Abstract:Thales Netherlands is the Thales Group’s Naval Centre for radar and combat management systems and is the largest defence company in the Netherlands. A growing trend to closer working relationships with the customer are performance-based contracts. For these kinds of contracts, Thales takes over all services for fixed costs. Thales will then manage the customer’s service and supply chain and the spare part stocks. The key performance indicator is the average system supply availability over a certain time period. The supply availability is the time the system is up divided by the total system’s operating time (uptime + downtime). The downtime of a radar system is measured as the time the system is waiting for a spare part. During a standing contract, it is possible that the attained system availability is lower than required and Thales may get a penalty. Due to the variation in the average availability, it is possible that Thales gets high bonuses and penalties. Besides this, operating hours of the ships vary and there may be more or less failures than expected (this affects the average availability). Estimating the results (bonuses and penalties) of a contract may become difficult and may have a large impact on the customer’s service perception. Currently, Thales increases the availability only by stocking extra spares. Since this is expensive, Thales wants to know which other tactical decisions there are that may increase the system performance. The goal of this research is to get insights into the impact of different tactical decisions on the system performance (availability and its variability). We focus on logistic parameters, which are stocking spare parts, repair throughput times, and order-and-ship times. With the goal in mind, we formulate the main research question as:“How can Thales use tactical decisions to improve the service contract performance at low costs, focusing on extra spare parts, decreasing order-and-ship times, and lowering mean repair throughput times?” It is possible that the original stock allocation is not optimal anymore since the demand or time parameters may have changed. The spare part allocation is optimally calculated using the spare part inventory tool “INVENTRI”. This tool is based on the Multi-Echelon Technique for Recoverable Inventory Control (METRIC). The multi-echelon, multi-indenture optimisation gives an optimal trade-off curve between spare part investment and average supply availability, in which maximising the average supply availability is seen as minimising the expected ship backorders. The impact of a tactical decision for an item is calculated as the expected ship backorder reduction per invested euro per year. We implemented the effects of the tactical decisions for an item in Excel. To improve the average availability, we defined three different greedy heuristics. The first heuristic looks at all items in the system and takes each time a tactical decision for a specific item, the second mainly focuses on availability killers, and the third takes each time the same tactical decision for all items together in the system. The third heuristic will never be optimal, since decisions are taken for items for which no backorder reductions may be attained. We use this heuristic only to see the impact of tactical decisions in general. In our case study we focus on a 3-echelon, 2-indenture supply network with six ships, one shore location, and one supplier (Thales). We used the heuristics to see whether the average availability, the variability, initial investments, and the robustness to changing annual Management Summary __________________________________________________________________________________ - iii - operating hours (AOH) can be improved with less costs than only stocking spare parts (INVENTRI solution). Based on the results, we draw the following conclusions: (1) “Heuristic one (tactical decisions over all LRUs) results in the lowest costs.” Using the first heuristic in the optimisation method results in the lowest annual costs, but the second heuristic leads to less tactical decisions for less items and has only slightly larger annual costs (6.3% versus 5.6% compared to INVENTRI). (2) “A combination of reducing time parameters and stocking extra spare parts leads to lower costs, a better variability, and a better robustness to changing AOH than only stocking spare parts.” Using the developed heuristics results in lower costs than the INVENTRI solution (only stocking spare parts). (3) “Lowering the gross mean repair throughput time at Thales is the best option to improve service contract performance.” Lowering gross repair throughput times has the largest impact on all aspects we looked at. (4) “Including subcomponents of expensive items with a high failure rate in the spare part allocation optimisation results in large savings in initial investments.” In our case, a reduction of 35 percent (€960,000) may be attained when subcomponents of the most expensive item (with the highest failure rate) are included in the spare part optimisation. (5) “Using a buffer in the net repair throughput time is disputable.” Since it seems that in the different processes of the net repair throughput time a buffer is used already, it is disputable whether including an extra buffer of two weeks is necessary. Finally, we give the following four most important recommendations for Thales to improve the service contract performance: (1) “Consider reductions in time parameters besides stocking spare parts.” Compared to stocking extra spare parts, using a combination of reducing time parameters and stocking extra spares is more cost effective, decreases the variability more than proportional, and make the system more robust to changes in annual operating hours. (2) “Analyse the impact of a tactical decision with the second heuristic.” Using the developed heuristics results in lower costs than the INVENTRI solution. Although the first heuristic results in the lowest costs, we recommend the second heuristic since this requires less tactical decisions and it is less difficult to use in practice. (3) “Better control the repair transaction process.” We showed that lowering the gross repair throughput times at Thales is the best option to improve the system performance. To catch up with variation in the different processes, a two-week buffer is used. However, decreasing the buffer already improves the performance and it has no extra costs. (4) “Always include subcomponents of expensive items with a high failure rate in the spare part allocation optimisation.” In this case, including subcomponents of only one item of this kind, results in a 35 percent reduction in the initial investment. When subcomponents are cheap, place plenty of them on stock and the gross repair throughput time can be reduced with the average waiting time for those subcomponents.
Item Type:Essay (Master)
Thales Netherlands B.V.
Faculty:BMS: Behavioural, Management and Social Sciences
Subject:85 business administration, organizational science
Programme:Industrial Engineering and Management MSc (60029)
Link to this item:
Export this item as:BibTeX
HTML Citation
Reference Manager


Repository Staff Only: item control page