The Bullwhip Effect Related to the Management of the Supply Chain - Research Paper Example

 International Purchasing, SCM Table of Contents I. Introduction II. The Bullwhip Effect and the Beer Game A. The Bullwhip Effect as Reflected in the Data, How the Bullwhip Affects the Beer Supply Chain Analyzing The Simulation Data Computing, Comparing Standard Deviation Figures for Inventory Data Along the Supply Chain B. Analysis- Causes of The Bullwhip in the Context of the Beer Game Simulation III. Mitigating the Bullwhip- Approaches and Strategies References I. Introduction This paper presents the simulation results for the beer game logistics simulation exercise. The paper uses an analysis of the simulation game data in order to contextualize the discussion of the bullwhip effect as it relates to the management of the supply chain. As required, the paper makes a critical evaluation of The Causes Of The Bullwhip Effect And The Way It Impacts The Management Of The Supply Chain. Then the paper goes on to explore various mitigating approaches and strategies to properly manage and mitigate the risks to the supply chain due to the bullwhip effect. It is important to understand the underlying dynamics and causes/effects of the bullwhip effect in the supply chain because the bullwhip effect introduces uncertainty, inefficiency, waste, and risks into the supply chain, and all of those imply larger costs, reduced profitability, wasted opportunities and a cascade of other negative consequences for a firm. As the class slides indicated, among the more pronounced consequences and effects of the bullwhip effect are reduced revenues, increased costs, decreased levels of quality, and reduced levels of service quality (Lin 2015). Moreover, the discussions noted that the bullwhip effect persists even in the absence of demand variability from the customer end, complicating management, and pointing to factors further along the supply chain that cause the bullwhip to arise (Cakanyildirim n.d.; Riemer 2012). On the other hand, among the factors that contribute to the occurrence and worsening of the bullwhip effect are the absence of coordinating, centralized data/information; increased lead times between orders; certain practices for forecasting demand, such as min-max; the practice of ordering by batches; and the practice of instituting discount practices and promotions practices, among others, that lead to fluctuations in prices. (Lee, Padmanabhan and Whang 1997). Meanwhile, the discussions also pointed out some ways to reduce the bullwhip intensity, and chief among the ways is leveraging more updated, centralised, data/information to fuel close to real time planning and to make sure that every step along the chain has the most recent customer demand data to inform its decisions and processes (Nienhaus 2002). Also among the ways to reduce the bullwhip are reductions in the lead times via the use of information and the use of cross-docking methods; working with vendors in managing the inventory; and to stabilize the demand via pricing strategies to reduce the swings in the supply chain (Cachon, Randall and Schmidt 2007). The rest of the paper discusses these and other strategies, as well as the underlying dynamics and causes of the bullwhip effect taking off from the results of the beer game simulation (Zavacka 2012; Reese 1995). Leveraging the literature, the paper fulfills the requirements and sets forth to explore the impact of downstream changes in demand to activities further upstream in the supply chain, as the cascade moves further away from the demand side or the customers upwards through the players in the chain (Wharburton 2004). II. The Bullwhip Effect and the Beer Game A. The Bullwhip Effect as Reflected in the Data, How the Bullwhip Affects the Beer Supply Chain Various definitions of the bullwhip effect trace the impact of distortions in information passed on to the rest of the supply chain going upstream through it as the root cause of the wild fluctuations and upswings in forecast figures for a product. In one definition for instance, the root cause is misinformation on demand, or what can be construed as the passing of distorted information on demand from the side of the customer propagating through the different players of the supply chain going upstream (Cakanyildirim n.d.). The same language of demand becoming distorted moving up the supply chain stream is present in discussions revisiting the bullwhip effect in more recent academic literature extending the notion of the bullwhip effect to the so-called green bullwhip effect, or the effect of changes in demand brought about by environmental considerations factoring into demand and production decisions (Lee et al. 2014). The same focus on information distortion and the elimination of such distortion through greater information sharing especially as one moves upstream along the supply chain is seen in literature coming out of Wharton relating to how behaviour shapes the bullwhip effect dynamics and how sharing information on inventory levels alleviates the bullwhip effect by flagging upstream players of the supply chain on the nature of demand downstream (Croson and Donohue n.d.). A Kellogg text on the other hand defines the bullwhip effect in terms of the propagation of magnifying effects of variability in demand due to members of the supply chain failing to properly coordinate their efforts. This lack of coordination or asynchronous efforts of course implies a lack of common data to work with, alongside other factors such as behaviour dynamics when different players in the supply chain try to make optimal decisions in the face of lack of data and uncertainties in the supply chain (Lariviere 2010). There are also references to the way the distortion in information and the resulting bullwhip propagating from the customer demand side all the way to the supplier side is amplified in proportion to the lead times from order to production and the lead times in both the flow of information and in materials along the chain (Chen et al. 2000; Chatfield et al. 2004). As those lead times get longer, so do the bullwhips amplify in strength, implying that one way to reduce the sting of the bullwhip and to reduce its amplification lies not only in being able to share information among players in the chain, but also in making sure that lead times in the flow of information are reduced, together with the total lead times from orders to the delivery of the final products (Metters 1997). Looking at the above discussion and sifting through the key insights from the literature relating to the nature of the bullwhip effect problem on the one hand and the dynamics of the beer game on the other, one can see that it is easy for the bullwhip to get out of hand in the latter game, owing to the presence of a considerable lead time, which in itself worsens the bullwhip and amplified distortions along the supply chain as discussed in the literature. Aside from the presence of considerable lead times there was a dearth of information among members that could have led to a more coordinated response to the fluctuations in the orders. Here behavioural dynamics factor in. Because players upstream are not aware of anything other than the orders coming into their part of the supply chain, and having the need to protect their own turf, they then overstate their orders that they pass up the supply chain stream, and so the distortions get out of hand and the bullwhip gyrates wildly, propagating the errors upstream (Croson and Donohue 2005). The lack of coordination moreover implies that in certain cases players upstream may misinterpret demand signals from downstream, and view attempts at reducing inventories as increased demand. The reality may be for instance that members downstream may just be trying to reduce inventories through special activities such as promotions and discounts in order to better manage inventory costs and reduce the burden on costs brought about by excess inventory. The flipside to this reality of course is if the beer game is set up in such a way that information sharing is perfect and the lead times for information sharing and in the fulfilment of orders is zero. In this case, there will be no bullwhip and there are no attempts by any of the players to overstate their orders and to overstock on needed materials and products. This is easier said than done but this is the ideal, and to the extent that the game conditions are not ideal, then we can explain why the bullwhip was observed and reflected in the recorded data, which will be analyzed and discussed in the following sections (Wu and Katok 2006). Analyzing The Simulation Data Looking at the simulation data in particular, we see a cascade of increasing inventories from the retailer to the wholesaler to the distributor and all the way upstream in the supply chain to the manufacturer from week 1 onwards, triggered in part by the lack of coordination among the different players in the chain. As the weeks progressed, moreover, we see from the data that the bullwhip took shape in the form of a cascade of ballooning inventories from the wholesaler down, all the way to week 5 when the surpluses in inventory flipped to become shortages and the orders shifted in magnitude by several factors. From week 1 to week 5 we see inventories being positive but the players continuing to place orders. On the other hand, by week 8 the magnitude of the shortages from the three latter players upstream in the chain had ballooned by several factors, when the level of orders had not changed by those same orders of magnitude. To cite some examples from the data, shortages from the manufacturer end ballooned to as much as 35 units in week 10, when the consumer orders remained at a tight range between 7 and 13 only. We see also that the large orders at the retailer end relative to actual demand from the customers caused wide disruptions in the forecasting of the players upstream in the chain, causing the manufacturer, the wholesaler and the distributor to register zero inventories and ordering way more than was actually required by the demand (Chatfield et al. 2004). From the point of view of coordination and synchronization, which would have been made possible by more timely sharing of information, the data reflects a chaotic dynamic where the individual participants in the chain reacted blindly to very severely constrained inputs from the rest (Metters 1997). Putting a cost to the bullwhip effects, by convention the cost of holding inventory is pegged at 0.5 while the cost of every unit of shortage is 1. From the data table, the total inventory across the supply chain at the end of the simulation, after week 12, is 214 and the total shortages is 146. The total inventory cost is therefore half of total inventory added to total shortages, or 253. This is the cost of the bullwhip effect to the firm. One can see too, from the table, that the bullwhip costs are amplified as one goes upstream along the chain, with the greatest costs being incurred by the manufacturer, with the greatest shortages and the greatest inventory levels at the end of week 12, and the costs dissipating as one moves downstream towards the customer. This is consistent with the literature on the bullwhip effect, where the observation is that the inefficiencies propagate through the supply chain starting from relatively small fluctuations in customer demand and small compensatory overstocking and over-ordering as defensive stances by players downstream. In the case of the retailer, small changes to over-order beyond what is being demanded at the consumer end propagates in terms of increasing costs through the different players along the supply chain. Looking at the inventory and shortages tallies, moreover, one can see that where the level of consumer orders had kept to a tight range, the shortages along the chain had ballooned, and this is due again to the inefficiencies in the forecasting and the ordering levels of the players downstream. The distributor then had to shoulder the inefficiencies in the ordering of the wholesaler and the retailer, and had to pay for its own overcompensation in orders too, by ending up ordering more than twice what the market was demanding. The manufacturer too ended up paying too much for overstocking on inventory that was essentially unneeded and mismatched with demand, while also suffering from incurring shortages costs that were several magnitudes higher than any other player downstream in the chain. One can see from these figures the operation of the bullwhip effect in the beer game simulation, and the costs highlight the effects and the extent of the bullwhip effect’s impact on firm operations and financial viability (Zavacka 2012; Reese 1995). Computing, Comparing Standard Deviation Figures for Inventory Data Along the Supply Chain Figures for the standard deviation for the shortages, orders and inventories for the different players along the supply chain also give an indication as to the extent of the variability of the numbers through the simulation, and also reflects the extent of the effect of the bullwhip on the different players. Using the standard deviation formula in Excel, this analysis was able to derive those standard deviation numbers. It is revealing that while the standard deviation for the consumer orders was just 1.8, the standard deviation figures for the inventory numbers upstream through the chain shows increasing dispersions in the values through the weeks, as reflected by higher standard deviation figures. The highest dispersions in the values for the inventories, shortages, and order quantities are to be found in the manufacturer data for those, while the lowest in the chain are to be found for data pertaining to the retailer. These standard deviation figures again are consistent with the expected finding from the literature that inefficiencies in the supply chain result in the propagation of bullwhip effects as one goes further away from the consumer and moves up through the supply chain all the way to the manufacturer. The table below reflects the simulation data, the computation for the costs, and the standard deviation figures for the different inventory data along the supply chain, summarizing the discussion above (Lee, Padmanabhan and Whang 1997): Data source: results of beer game simulation 2015 B. Analysis- Causes of The Bullwhip in the Context of the Beer Game Simulation As discussed above, the scenario for the beer game is less than ideal for the maintenance of a supply chain that has no bullwhip or has minimal bullwhip effects. The ordering process was done in silos, with the different players operating with limited information and not undertaking any concerted effort to tame any inefficiencies along the chain. Moreover the lead time between ordering and orders fulfilment was significant, and without any informed cues from other players about what was happening at the level of the consumer, with little to no information on what is happening elsewhere in the chain, the players had to make the best possible decision to tame their own individual costs and to serve all orders as best they can. As the literature notes, even with well-managed supply chains that do their best to eliminate lead times and to share as much information as possible, the bullwhip effect still materializes due to other, behavioural and consumer-tied dynamics. It is no surprise then that in the haphazard way in which the ordering process was done for each of the players in the chain, that the bullwhip would show up with severe effects propagating through every stage of the supply chain. Moreover, the literature notes that where the bullwhip effect is observed, the lengthening of the lead times for orders fulfilment from the top of the supply chain is associated with the worsening of the effects of the bullwhip. The presence of considerable lead times in the beer game simulation therefore can be construed as a marker for expected large bullwhip effects in the beer supply chain. The results of the data analysis confirm this. The results of the simulation therefore are in keeping with the expectations from the literature with regard to the effect of long lead times and non-coordinated activities and no information sharing among players on the presence and extent of the bullwhip effect. Simply put, the scenario conditions and the observed bullwhip effects in the simulation data confirm the expectations from the academic literature (Croson and Donohue n.d.; Zavacka 2012). III. Mitigating the Bullwhip- Approaches and Strategies The previous discussion and the survey of the literature reveals insights into how to mitigate the bullwhip. Essentially, the key is to enable information sharing among as many players as possible and to allow them to coordinate and collaborate on ways to shorten lead times, better forecast demand, and coordinate production runs and even marketing strategies and trade strategies (Lin 2015). From the very bottom of the supply chain, the customer-facing players in the chain can share information on demand, and activities to stimulate demand, as close to real time as possible. Real-time consumption data can then be fed into the supply chain information system to fulfil new orders as close to the actual demand numbers as possible, eliminating the need for guess work in forecasting. On the other end, shortening lead times for orders fulfilment to as much as possible will translate to lower inventory levels, fewer shortages, and therefore lower costs overall. The ideal scenario is provided by the literature, and includes both an optimal orders fulfilment process with the shortest lead time possible, and the pervasive sharing of information and the closest possible collaborative relationship among members of the supply chain, from the retailer all the way to the manufacturer (Chatfield et al. 2004; Metters 1997). The literature provides specific technologies and implementation frameworks for essentially achieving pervasive information sharing, collaboration, and the shortening of lead times as discussed above (Chatfield et al. 2004). EDI technologies for instance allow for the real time capture of transactions at the consumer level, which can then be fed into a central information system that all players in the chain can utilize to forecast demand and plan for fulfilment in concert with the other players in the chain (Lin 2015). Manufacturers can also make use of real-time consumption data to coordinate production activities with its own suppliers. The key among all these technologies and frameworks is essentially that compression of lead times and the effective leveraging of live data that is as close to ground zero for consumers as possible, to allow for the freshest inputs to guide the activities of all players in an optimized supply chain for beer in this case and for any other product for that matter (Croson and Donohue 2005). 1 References Cakanyildirim, M. (n.d.). Bullwhip Effect. University of Texas at Dallas. [online]. Available at: http://www.utdallas.edu/~metin/Or6366/Folios/SourcePartner/scbullwhip.pdf [accessed 2/20/2015]. Cachon, G., Randall, T. and Schmidt, G. (2007). In Search of the Bullwhip Effect. Manufacturing & Service Operations Management/ Wharton School at the University of Pennsylvania. [online]. Available at: http://opim.wharton.upenn.edu/~cachon/pdf/bwv2.pdf [accessed 2/20/2015]. Chatfield, D. et al. (2004). The Bullwhip Effect- Impact of Stochastic Lead Time, Information Quality, and Information Sharing: A Simulation Study. Production and Operations Management Vol. 13 No. 4, pp. 340-353. [online]. Available at: www.researchgate.net/profile/Terry_Harrison2/publication/227541717_The_Bullwhip_EffectImpact_of_Stochastic_Lead_Time_Information_Quality_and_Information_Sharing_A_Simulation_Study/links/0c9605183362d3d2e6000000.pdf [accessed 2/20/2015]. Chen, F. et al. (2000). Quantifying the Bullwhip Effect in a Simple Supply Chain: The Impact of Forecasting, Lead Times, and Information. Management Science Vol. 46 No. 3, pp. 436-443. [online]. Available at: http://www.diku.dk/hjemmesider/ansatte/pisinger/production/SimchiLevi_BullwhipEffectSupplyChain.pdf [accessed 2/20/2015]. Croson, R. and Donohue, K. (n.d.). Behavioral Causes of the Bullwhip Effect and the Observed Value of Inventory Information. SystemDynamics/Wharton School/The Carlson School. [online]. Available at: http://www.systemdynamics.org/conferences/2003/proceed/PAPERS/234.pdf [accessed 2/20/2015]. Croson, R. and Donohue, K. (2005). Upstream versus downstream information and its impact on the bullwhip effect. System Dynamics Review Vol. 21 No. 3, pp. 249-260. [online]. Available at: http://www.researchgate.net/profile/Rachel_Croson/publication/227657683_Upstream_versus_downstream_information_and_its_impact_on_the_bullwhip_effect/links/0c96051c462802afa1000000.pdf [accessed 2/20/2015] Lariviere, M. (2010). Cracking the bullwhip. Kellogg Insight Presents The Operations Room. [online]. Available at: https://operationsroom.wordpress.com/2010/01/29/cracking-the-bullwhip/ [accessed 2/20/2015]. Lee, S. et al. (2014). The green bullwhip effect: transferring environmental requirements along a supply chain. International Journal of Production Economics Vol. 156, pp. 39-51. [online]. Available at: http://www.sciencedirect.com/science/article/pii/S0925527314001662 [accessed 2/20/2015]. Lee, H., Padmanabhan, V. and Whang, S. (1997). The Bullwhip Effect in Supply Chains. MIT Sloan Management Review Magazine. [online]. Available at: http://sloanreview.mit.edu/article/the-bullwhip-effect-in-supply-chains/ [accessed 2/20/2015]. Lin, Y. (2015). Session 10: Logistics Simulation- The Beer Game (Presentation Slides). BUSI 1320, University of Greenwich. Metters, R. (1997). Quantifying the bullwhip effect in supply chains. Journal of Operations Management Vol. 15, pp. 89-100. [online]. Available at: www.researchgate.net/profile/Richard_Metters/publication/222505541_Quantifying_the_bullwhip_effect_in_supply_chains/links/00b7d537cf4ae16af0000000.pdf [accessed 2/20/2015]. Nienhaus, J. (2002). What is the Bullwhip Effect caused by? Swiss Federal Institute of Technology (ETH) Zurich. [online]. Available at: http://www.beergame.lim.ethz.ch/Bullwhip_Effect.pdf [accessed 2/20/2015]. Reese, J. (1995). Whang and Lee: Eliminating the Bullwhip Effect in Supply Chains. Stanford Graduate School of Business. Available at: http://www.gsb.stanford.edu/insights/whang-lee-eliminating-bullwhip-effect-supply-chains [accessed 2/20/2015]. Riemer, K.(2012). Bullwhip Effect. The Beergame Portal. [online]. Available at: http://www.beergame.org/the-game/bullwhip-effect [accessed 2/20/2015]. Wharburton, R. (2004). An Analytical Investigation of the Bullwhip Effect. Production and Operations Management Vol. 13 No.2, pp. 150-160. [online]. Available at: http://www.poms.org/journal/2004-02-warburton.pdf [accessed 2/20/2015]. Wu, D. and Katok, E. (2006). Learning, communication, and the bullwhip effect. Journal of Operations Management Vol. 24 No. 6, pp. 839-850. [online]. Available at: http://www.utdallas.edu/~emk120030/learn_post.pdf [accessed 2/20/2015]. Zavacka, V. (2012). The bullwhip effect and the Great Trade Collapse. European Bank for Reconstruction and Development. Available at: http://www.ebrd.com/downloads/research/economics/workingpapers/wp0148.pdf [accessed 2/20/2015]. Read More