Wednesday, August 14, 2013

Installed Base Forecasting Can Cut Downtime in the Freight and Logistics Industry

Reactive Inventory Management Increases Delays and Costs
Shipping News Feature

GERMANY – WORLDWIDE – Institutions such as the Kuehne Logistics University (KLU) in Hamburg can play a valuable role in determining ways to improve the supply chain simply by offering the chance for study into the complexities involved in the flow of freight from manufacturer to end user. Dr. Çerag Pinçe, KLU’s Assistant Professor of Operations and Supply Chain, has been researching downtime, one of the scourges of any operator in the field, but his work has particular relevance for any major logistics supplier.

The KLU funded research project believes that it may have found the solution to the expensive problem in something called ‘Installed Base Forecasting’. Currently many companies hedge against the potential problems of downtime by overinvesting in stocks of service parts to ensure their equipment stays up and running. As any stockholder knows, maintenance of such a system is notoriously difficult, since it is impossible to predict how many spare parts will be needed and when, with a great degree of accuracy.

Dr. Pinçe calls this method of stock control ‘black-box forecasting’ and concludes that the reactive system used in such modules depends on the quality and quantity of historical data, which is typically intermittent and scarce. An alternative approach is to be proactive by observing the changes in the installed base, then integrating this information into the forecasting management system to predict future demand shifts, this approach is what he refers to as installed base forecasting.

In this case the demand for service parts is correlated to individual equipment units by first forecasting the changes in the number, age, location, etc. of these units, and then relating service parts demand to these changes. In summary, it uses black-box forecasting methods to estimate the individual demand rates for parts and then adjusts the estimates to the size of the installed base using actual stock figures. By using more information in forecasting the machine level demand rate, and decoupling it from aggregate level demand, yields much better estimates in comparison to using black-box forecasting methods alone.

A side benefit of this approach is that it provides feedback on real failure rates per machine to the engineering department. For service providers, the use of installed base information can be particularly instrumental in foreseeing large shifts in the demand rate and initiating proactive measures (e.g., running the stocks down or up before the shift occurs), since this approach requires improved monitoring of the customers and the equipment under service contract.

Many service companies of course are already collecting data on their installed base to predict short term parts requirements, or for making other operational decisions. However, it seems that they neither direct this information towards service contract expiration predictions nor entirely integrate it in their inventory management systems. The doctor concludes that installed base forecasting appears to be a promising method that may lead to improved spare parts forecasting. The effort put into establishing such an installed base forecasting system would prove worthwhile, especially when small installed bases of expensive equipment or large installed bases of standardised, cheap equipment are under consideration.

Dr. Çerag Pinçe is Assistant Professor of Operations and Supply Chain Management at the Kuehne Logistics University. He received his PhD in Management from Erasmus University, Rotterdam and MS in Industrial Engineering from Bilkent University, Ankara. Before joining the KLU he worked as a Postdoctoral Research Fellow at the College of Management at Georgia Institute of Technology, Atlanta.

The KLU, in the throes of setting up in HafenCity, Hamburg (the rough equivalent of London’s Docklands development) currently offers two Master’s programs in Global Logistics and Management, one part-time Executive Master in Leadership & Logistics and a PhD programme. From autumn 2013 onwards a Bachelor in Management will be introduced and parties interested in the university can type KLU into the News Search box at the top of this page.  

Photo: Dr. Çerag Pinçe