In the automotive industry, the high process complexity becomes an important issue because of the increased number of product and process variants demanded by the customers. To avoid quality defects in assembly and losses in such a complex manufacturing environment, new predictive support systems are required. This study aims to develop a multiple attribute decision support system (DSS) for the prediction and quantification of the risk of failures on the workstations of a leading Turkish automotive manufacturing company.
Initially, the factors affecting the failures in workstations and the attributes to evaluate the factors are identified. Subsequently, the relations among the attributes are specified and priorities of them are calculated. Finally, the risk of failures is calculated and tested in a pilot study and validated with real production data.
To the best of authors’ knowledge, this is a unique study that computes the risk scores on the workstations via DSS. The DSS has various advantages for improvements of the manufacturing quality: the risk of failures can be detectable and comparable, the effect of changes in the design of new workstations can be observed. Stations that have medium or high complexity scores demonstrated strong correlation with failure rates. A sensitivity analysis is conducted to predict the effect of improvement actions on the riskiness of the workstations.
High level of production complexity becomes a crucial issue for companies that use various production processes. Considering this fact, it is a requirement for companies to observe and monitor the risk factors, especially in the assembly lines to be able to eliminate failures derived from complexity. Accordingly, to measure risk scores of the workstations in the assembly lines, a decision support for companies aids executives to manage the complexity level in a reliable and effective way. In this study, the authors develop such a DSS for TOFAS, a leading Turkish automotive company. The proposed DSS is verified and applied through a pilot study on a specific basic production unit. A sensitivity analysis is also conducted to see the effects of potential improvements on the risk scores. Additionally, the trend of risk scores for the stations can also give valuable information for tracing the changes in the time horizon. The proposed DSS also enables an opportunity for the executives in their decision of design processes of new production lines by allocating limited resources in an appropriate way based on the risk scores of possible workstations. The proposed DSS is the first and unique proactive failure prevention model developed in a Fiat Chrysler Automobiles (FCA) plant across the world. TOFAS executives also plan to introduce and enlarge the usage of the model to other FCA plants. It may also be possible to apply the model to other assembly lines in any sector. Another plan of the executives of TOFAS is developing a software, which manages each parameter, to constitute data to the DSS to run this system more instantly and effectively. Moreover, they can take integration actions of the software with world-class manufacturing problem management system that is currently in use in TOFAS.
This study is a research and development project conducted through the collaboration of ITU and TOFAS, and it was funded by TOFAS. The authors are thankful to the managers and specialists of TOFAS for their invaluable cooperation during the structuring and modeling phases of the proposed model. The authors also owe a great deal of thanks to Mine Isik for her invaluable contribution for the identification of risk attributes and Ali Ali for his assistance during the pilot study.
Unver, B., Kabak, Ö., Topcu, Y.I., Altinisik, A. and Cavusoglu, O. (2020), "A decision support system for proactive failure prevention: a case in a leading automotive company", Journal of Enterprise Information Management, Vol. ahead-of-print No. ahead-of-print. https://doi.org/10.1108/JEIM-09-2019-0264Download as .RIS
Emerald Publishing Limited
Copyright © 2020, Emerald Publishing Limited