LT variance or LT mean reduction in supply chain management: Which one has a higher impact on SC performance? S. Kamal Chaharsooghi à , Jafar Heydari Industrial Engineering Department, School of Engineering, Tarbiat Modares University, Tehran, Iran article info Article history: Received 1 August 2007 Accepted 3 December 2009 Available online 24 December 2009 Keywords: Lead time variance Lead time mean Simulation Canonical correlation Multivariate analysis abstract Lead time (LT) is an inseparable factor of any supply chain (SC). Lead time uncertainty is known as a type of supply uncertainty that affects ordering policies, inventory levels, and product availability level. There are two main strategies concerning SC uncertainties: adapter and shaper strategies that can be used in environments with LT uncertainty. As for LT mean, it can be reduced by improving the performance of upstream members. In this paper, relative importance of these two parameters to be reduced by these strategies is determined by investigating the effects of both LT mean and LT variance on supply chain performance indices. Simulation and multivariate models have shown that LT variance has a stronger impact on SC performance measures (including Bullwhip effect, holding inventory, stock-out size and number of stock-outs). This study can help managers in (1) rebalancing of these two parameters by applying a proper investment strategy; (2) suitable selection of service providers in transportation based on LT parameters. & 2009 Elsevier B.V. All rights reserved. 1. Introduction Supply chain consists of all the stages in fulfilling a customer request (Chopra and Meindl, 2006). These stages are to supply, manufacture, and retail, as well as to transport, warehouse, and distribute. Each product should pass through all these stages in order to be available to the end-consumer. One of the interests of managers is to eliminate the waste throughout the supply chain, resulting in development of strategies, such as JIT. However, due to the environment complexities, attaining such elimination becomes more difficult. Delivery lead time is one of the main factors that affects supply chain performance and makes the planning process more complex. From inventory theory, it is well known that delivery lead time of the upstream affects the inventory parameters of the downstream, such as inventory position, product availability level and ordering decisions. Thus, one may conclude that LT affects overall SC performance through the inventory and ordering systems. Lead time is defined as the time interval between placing an order and its reception. Based on this definition, lead time is composed of various parts including (1) information delays in receiving an order by the upstream, (2) time of order processing by the upstream, and (3) transportation time. Each part is affected by SC member performance and external factors. Low supply performance results in increased lead time. For example, if a supplier has a low order processing performance, the time of order reception by the downstream will be longer, leading to a higher LT mean. The supplier’s fast response to the received order can reduce the LT mean. External factors refer to environmental uncertainties that cause uncertainty in lead time. For example, transportation time is highly sensitive to the environmental uncertainties, such as weather conditions; thus, the variance of order reception time will be high even by using the same transportation mode (e.g. road). There are two main strategies concerning uncertainties: shaper and adapter strategies. Shaper strategies try to reduce the uncertainty level, while adapter strategies accept the uncertainty but try to reduce its effect by adapting SC members accordingly. Reducing LT and its uncertainty is considered as an investment (Bookbinder and C - akanyildirim, 1999; Ryu and Lee, 2003). It is then possible to reduce both LT mean and variance using a proper investment strategy. Each of these investment strategies has its own characteristics of cost and outcome. Thus, it is crucial to select the most suitable investment strategy based on profitability measures. On the other hand, it is possible in some cases to trade-off LT mean and variance. For example, consider a case in which we have two transportation service providers to select between one with less time variance but longer transporta- tion time, the other with more time variance but shorter transportation time. Preferring one alternative over the other needs a trade-off. The decision in this case needs an under- standing of the effects of LT on performance measures of the chain. This paper helps decision makers to select between investment strategies by determining the relative priority of LT mean and ARTICLE IN PRESS Contents lists available at ScienceDirect journal homepage: www.elsevier.com/locate/ijpe Int. J. Production Economics 0925-5273/$ - see front matter & 2009 Elsevier B.V. All rights reserved. doi:10.1016/j.ijpe.2009.12.010 à Corresponding author. Tel./fax: + 98 21 44209944. E-mail addresses: SKCH@modares.ac.ir (S.K. Chaharsooghi), Heidarij@modares. ac.ir (J. Heydari). Int. J. Production Economics 124 (2010) 475–481