A fuzzy approach for total productive maintenance performance measurement in manufacturing systems

Total Productive Maintenance (TPM), which recognized as a strategic maintenance technique, has been widely and successfully implemented in many organizations. In this context, evaluation of TPM performance can make a great contribution to organizations in advancing their manufacturing operations. Therefore, this thesis aims to develop a new framework for the performance measurement of TPM based on quantitative and qualitative performance indicators. Within the scope of this thesis, the proposed TPM performance measurement system (TPM PMS) is divided into four phases namely design, evaluation, implementation, and review. In the design phase, novel performance indicators having impact on TPM performance are identified and analyzed. In the evaluation phase, these indicators are evaluated using a fuzzy multiattribute decision making (FMADM) method improved on the basis of fuzzy arithmetic and ranking. Moreover, the improved method is compared with the most popular FMADM methods in the literature and its applicability and reliability are determined by carrying out sensitivity analysis. In the implementation phase, TPM performance is measured with novel performance indicators using fuzzy data envelopment analysis (FDEA). In this context, different generalized fuzzy data envelopment analysis with assurance regions models are proposed in the presence of desirable and undesirable inputs and outputs. Thus, the proposed models make a significant contribution into TPM literature. In the review phase, TPM performance should be monitored periodically, and preventive and predictive decisions or actions should be taken if it is needed. Finally, the proposed TPM PMS is implemented in an international manufacturing company operating on automotive industry.