Enhancement of a statistical life-cycle cost model for aero engines with support of service data analysis

This Master´s thesis covers the improvement of a life-cycle cost (LCC) model to predict the shop visit costs of Rolls-Royce BR725 aero engines. Based on estimates and previous engines maintenance data, this model utilises an internally developed tool (DMTrade) to predict the average time on wing from entry into service to the first shop visit and the respective costs. The first part of the analysis was based on the optimization of the model. Initially, DMTrade was modified in order to display the costs relative to the specific case of study, which was the first proactive engine life management (PELM) shop visit. Using a previous BR725 model as a starting point, the main engineering and cost assumptions were gradually updated and the impact of each modification was assessed. Once a complete and up-to-date version was achieved, a new model was implemented to summarize the latest inputs in a more efficient and organized structure: this contains the most recent assumptions in terms of removal drivers distributions, scrap and repair rates and parts prices. The second part of the study focused on the validation of the life-cycle cost model using Rolls-Royce BR710 aero engine in-service fleet data, to ensure the full applicability and correspondence of the assumptions made. For this purpose, scrap rates inputs were extracted from an in-service fleet database, Engine Network. To complete the validation model, the latest repair rates, Life Limited Parts (LLP) lives and costs assumptions were taken from the most recent BR710 internal documents. The total cost obtained was compared to a target value determined from a sample of BR710 PELM shop visits data. This assessment showed that the model implemented can be used as a trustable tool for shop visit costs forecastings, setting a starting point for future studies in the field of life-cycle cost analysis. Future investigations are suggested in order to improve the precision and the quality of the cost predictions.