Title: A service level agreement baselining methodology for non-normal characteristics using the Pearson distribution
Authors: Boby John; S.M. Subhani
Addresses: SQC & OR Unit, Indian Statistical Institute, 8th Mile, Mysore Road, Bangalore, 560-059, India ' SQC & OR Unit, Indian Statistical Institute, Street # 8, Habsiguda, Hyderabad, 500-007, India
Abstract: Many organisations use outsourcing to achieve cost advantage and focus on core processes. The quality of the services provided by the outsourced partner plays a significant role in the firm's overall performance. The service level agreement (SLA) baselining is the process of finalising the performance level expected from the outsourced partner. Generally, performance characteristics like cycle time, turnaround time, handling time, productivity, etc. are baselined. Many of the aforementioned metrics may not be normally distributed. In this paper, the authors discuss the cases of baselining non-normal performance characteristics. The methodology is to fit a general system of distributions namely the Pearson family of distribution and identify the SLA value using the best fit Pearson type of distribution. The methodology is demonstrated through two case studies and results are found to be very encouraging. The methodology can be used for baselining of non-normal performance characteristics of various service industries.
Keywords: service level agreement baselining; process capability analysis; non-normal characteristics; Pearson family of distributions; R package.
International Journal of Industrial and Systems Engineering, 2021 Vol.37 No.2, pp.222 - 240
Received: 27 Nov 2018
Accepted: 28 Jun 2019
Published online: 08 Feb 2021 *