Framework for Improving Spare Parts Availability Using Big Data Analytics

Big data has brought about significant changes to various industries in the last few years, resulting in the expectation that it might transform the way supply chain continues to develop (Waller & Fawcett, 2013). Main characteristics of big data are volume, velocity and variety (Sanders, 2014; Waller & Fawcett, 2013; Wang et al., 2016). But, these qualities alone are not enough to deliver meaningful insights (Sanders, 2014), it is the combination of analytics and big data which provides real value to supply chain decision making (Baines and Lightfoot, 2013). There is a need to better understand the framework for big data management and the benefits which big data can bring (Dutta & Bose, 2015). Especially in big data environment, the managers should have both deep understanding of the business as well as have perspectives on utilizing the data for business decisions (Waller & Fawcett, 2013).

Currently, many manufacturing organisations are moving towards servitization to add greater value for their customers (Johnson and Mena, 2008; Neely, 2008). Timely spare parts provisioning is one key element of the servitized businesses for such companies (Saccani et al., 2014). Providing performance guarantees and hence minimal productivity losses for customers require deeper understanding of failure rates of nonrepairable spare parts. Characteristics of such parts are intermittent and erratic demand, large variety, frequent obsolescence (Diaz, 2003), critical to operations (Huiskonen, 2001), and slow moving (Dekker et al., 2013), But, often shortage of historical data makes inventory management of spare parts more complicated (Huiskonen, 2001; Stefanovic, 2015). Hence, big data which can capture the real time usage of such parts has high potential for better failure prediction, spare parts planning and availability and higher productivity for customers. By analyzing such big data, a company will be able to predict when the failure is likely to occur (Baines and Lightfoot, 2013). The appropriate use of big data can reduce unplanned downtimes at manufacturing facilities (Munirathinam & Ramadoss, 2014) and the likelihood of the breakdowns. Remote monitoring technology can be used to capture real time performance of parts (Grubic and Peppard, 2016). While Grubic and Peppard (2016) provide the enablers and barriers of using remote monitoring technology for servitization, there is lack of a comprehensive framework and implementation plan of how companies with large variety of non-repairable spare parts can enhance performance of their service business using big data. This research tries to address this gap with the following objectives:

• To identify the key parameters for predicting potential failures of spare parts.
• To outline the data collection and analysis processes for spare parts demand planning.
• To develop a framework for big data implementation for spare parts planning and identify the benefits which can be derived from it.

A single in-depth case study is an appropriate unit of analysis for the research as dangers of making crosssectoral generalisations regarding servitization strategies (Johnstone et al., 2009) and more so for big data analytics for servitization which may be highly context-specific.

To fulfill the objectives of this research, we chose a company which has decided to invest in smart-parts to differentiate themselves from the competitors and to improve their service business. Smart-parts is an approach for condition monitoring of parts with technology that users sensors. Any information from the sensors about potential failures can potentially allow the company to improve planning and increase their spare parts availability. However the company faces significant challenges in introducing those smartparts. One of the challenges is identifying which data should be collected and how to plan the inventory using this data. The other challenge is how to manage the process for big data implementation in the company. As the company had numerous types of spare parts, we decided to focus on three spare parts due to their substantial contribution to the service revenue and since from a customer’s point of view these parts are more suitable for data capture by sensors. By studying the characteristics of these parts, the current maintenance and spare parts planning practices, a framework for utilizing big data for spare parts planning in the organization utilizing both historical and real time condition monitoring data is developed.

The parameters for predicting failure are identified for the chosen parts based one literature and knowledge of experts. Then three types of data which should be collected for example installed base data, condition monitoring data and event data are specified and finally some practical considerations for implementation are outlined. This implementation would help improve spare parts planning and consequently improve the performance of the service business. The overall framework is shown below:

References

Baines, Tim and Lightfoot , Howard, W.(2013). Servitization of the manufacturing firm. International Journal of Operations & Production Management 34(1): 2 – 35

Dekker, R., Pinçe, Ç., Zuidwijk, R., & Jalil, M. N. (2013). On the use of installed base information for spare parts logistics: A review of ideas and industry practice. International Journal of Production Economics, 143(2), 536-545.

Diaz, A. (2003). Modelling Approaches to Optimise Spares in Multi-echelon Systems. International Journal of Logistics, 6(1-2), 51-62.

Dutta, D., & Bose, I. (2015). Managing a Big Data project: The case of Ramco Cements Limited. International Journal of Production Economics 165, 293-306

Grubic, T., & Peppard, J. (2016). Servitized manufacturing firms competing through remote monitoring technology: An exploratory study. Journal of Manufacturing Technology Management, 27(2), 154-184.

Huiskonen, J. (2001). Maintenance spare parts logistics: Special characteristics and strategic choices. International Journal of Production Economics, 71(1-3), 125-133

Johnson, M., & Mena, C. (2008). Supply chain management for servitised products: a multi-industry case study. International Journal of Production Economics, 114(1), 27-39.

Johnstone, S., Dainty, A., & Wilkinson, A. (2009). Integrating products and services through life: an aerospace experience. International Journal of Operations & Production Management, 29(5), 520-538.

Munirathinam, S., & Ramadoss, B. (2014). Big data predictive analtyics for proactive semiconductor equipment maintenance. Paper presented at the Big Data (Big Data), IEEE International Conference on. 27-30 Oct. 2014, Neely, A. (2008). Exploring the financial consequences of the servitization of manufacturing. Operations Management Research, 1(2), 103-118.

Saccani, N., Visintin, F., & Rapaccini, M. (2014). Investigating the linkages between service types and supplier relationships in servitized environments. International Journal of Production Economics, 149, 226-238.

Sanders, N. R. (2014). “Big” Data Driven Supply Chains. Upper Saddle River, NJ: Pearson Education (US). Schlegel, Stefanovic, N. (2015). Collaborative Predictive Business Intelligence Model for Spare Parts Inventory Replenishment. Computer Science and Information Systems, 12(3), 911-930.

Waller, M. A., & Fawcett, S. E. (2013). Data science, predictive analytics, and big data: a revolution that will transform supply chain design and management. Journal of Business Logistics, 34(2), 77-84.

Wang, G., Gunasekaran, A., Ngai, E. W. T., & Papadopoulos, T. (2016). Big data analytics in logistics and supply chain management: Certain investigations for research and applications. International Journal of Production Economics, 176, 98-110

Respected Authors of this White Paper

Julija Ivcenko

Operations and Management Engineering Graduate, Aalborg University, Copenhagen

julijaivcenko@gmail.com, +4550362449

Atanu Chaudhuri

Center for Industrial Production, Aalborg University, Copenhagen

Atanu@business.aau.dk, +45 9940 3029