Dealing with Segmented Supply Chains

Stephen DeAngelis

July 14, 2011

Supply chain risk management guru Daniel Stengl notes that more and more companies are developing product lines for different markets and wonders “if it is ok just to use the same supply chain strategy for all those products”? [“Managing Supply Chains with multiple Pipelines,” Supply Chain Risk Management, 10 January 2011] Drawing from a paper written by James Aitken, Paul Childerhouse, Martin Christopher & Denis Towill [“Designing and Managing multiple Pipelines,” Journal of Business Logistics, 26 (2), 73-95, 2005], Stengl believes the authors “make a convincing argument against this approach.” In the Abstract for their paper, the authors write:

“There is now a growing recognition that supply chains should be designed from ‘the customer backwards’ rather than from ‘the company outwards’. If such a view is accepted then the implication is that since the organisation will likely be serving multiple markets or segments there will be the need to design and manage multiple ‘pipelines’ to serve those different customers.”

These academics are joined by numerous supply chain analysts who also believe that supply chains should be designed from the outside in. As I wrote in a previous post, “noted supply chain analyst Lora Cecere writes, ‘We will never achieve supply chain signal synchronization working from the inside out. Instead, we must start with the requirements of the customer’s customer and work backwards translating the requirements to the suppliers’ supplier’ [“Dancing with a White Elephant,” Supply Chain Shaman, 25 April 2010]. Aitken and his colleagues talk about “pipelines” but others have used terms like “multi-channel” or “segmented” supply chains. Segmentation often refers to the supply chain while multi-channel often refers to sales outlets. Aitken and his colleagues based their work on a four-year case study of a lighting company that produces a range of products for diverse markets. After reading the study, Stengl concludes:

“Pipelines are a useful way of differentiating supply chains and selecting the right strategies depending on the market and product properties. Aitken et al. describe a normative scenario. In practice differences in supply chains are often due to missing integration of different parts of a company (e.g.. after a merger). But especially in those cases it is necessary to develop an approach for reintegration without losing sight of the different needs of the customers.”

Supply chain analyst Trevor Miles insists that mass customization is not going to go away. [“Visibility, the antidote to supply chain opacity,” The 21st Century Supply Chain, 2 May 2011]. That means that multi-channel supply chains are also going to be a fixture of supply chains as we move forward. David Rae, editor of Procurement Leaders, insists that growing complexity is “making legacy supply chains … less attractive every year.” By legacy supply chains, Rae means one-size-fits-all supply chains. An article published in the McKinsey Quarterly asserts that building the supply chain of the future “means ditching today’s monolithic model in favor of splintered supply chains that dismantle complexity, and using manufacturing networks to hedge uncertainty” [“Building the Supply Chain of the Future,” by Yogesh Malik, Alex Niemeyer, and Brian Ruwadi, January 2011]. This advice is similar to that coming from Gartner that states, “Remember that one size does not fit all. Define how many supply chain types you have and design a customized response for each” [“The AMR Supply Chain Top 25 for 2010,” by Kevin O’Marah and Debra Hofman, Gartner, 2 June 2010] O’Marah and Hofman insist that the best supply chains will be built from the “outside in, starting with your customers and working back through your trading-partner network to design a profitable response.” Referring to the McKinsey study, Rae writes:

“The paper spends a lot of time describing the challenges … and reaches a fairly accurate conclusion: ‘In such a world, the idea that companies can optimise their supply chains once — and for all circumstances and customers — is a fantasy’.”

There is a verse from the Old Testament that reads: “What has been will be again, what has been done will be done again; there is nothing new under the sun.” [Ecclesiastes 1:9] Simon Ellis, from IDC Manufacturing Insights, believes the sentiments expressed in that verse are true when it comes to supply chain segmentation. He writes:

“Supply Chain segmentation is not new, despite recent articles to the contrary, as manufacturing companies have explored ways to manage parts of the supply chain differently for the varied requirements of product lines or categories for decades. … New life is being breathed into an old concept.” [“Supply Chain Segmentation is Back!” Manufacturing Value Chain, 8 April 2011]

Ellis agrees with Rae that increased complexity is behind the renewed interest in supply chain segmentation. He writes:

“As we move into 2011, the business landscape has changed. The demand side of the supply chain is far more volatile and unpredictable, with forecast accuracy becoming even more of a challenge than it has historically been, and the conversation is starting to move to topics like fast planning and responsiveness. The supply side of the supply chain is also experiencing some seismic shocks (one literal, others figurative) that are causing manufacturing companies to relook at their supply networks through the lens of risk management and lead-time assessment. Whether these changes, on both the demand and supply side of the supply chain, prove to be short or long lived, there is little question that fundamentally, the complexity level that supply chain organizations are asked to manage has increased significantly over the last few years. Driven by complexity, and an increasingly diversified manufacturing industry (where many companies have a broad range of product categories, often crossing B-to-B/B-to-C lines), segmentation is a way to address complexity where and when it is necessary, but reject it where it is unnecessary. At its core, supply chain segmentations is quite simple, in that it is about designing supply chains at a product category or family level based on the unique needs of those product groupings.”

As Ellis noted above, “segmentation is a way to address complexity where and when it is necessary, but reject it where it is unnecessary.” He goes on to provide a few suggestions about how to determine “where segmentation lines might be drawn.” They are:

“● Short versus long-lead time products, where the former hold some level of fashion element and the latter do not

“● High versus low gross-margin products, where the former may obligate higher service levels (and a greater tolerance for returns) than the latter.

“● High versus low innovation churn categories, where the former may require a more collaborative supply network than the latter that is characterized more as ‘set it and forget it’.

“● SKU proliferation, particularly with ‘flavors and fragrances’ may drive higher levels of postponement for some product categories (consumer products) than others (chemicals).

“● Global versus regional demand products, where the former will require global fulfillment capabilities and the latter will not.”

With increased complexity now defining the modern supply chain, Srinivas Ambatipudi asks, “Are Supply Chain Planning systems (both demand and fulfillment) in a state where they can predict and plan the increasingly volatile demand?” [“Is there a need for Next Generation Supply Chain planning products?” Supply Chain Management, 31 May 2011] It makes sense that planning systems must change as supply chains segment. Ellis’ suggestions about where the lines need to be drawn between segmentation and non-segmentation could also be used to determine where different planning systems must be used. McKinsey’s Malik, Niemeyer, and Ruwadi insist that “many global supply chains are not equipped to cope with the world we are entering.” They explain:

“Most were engineered, some brilliantly, to manage stable, high-volume production by capitalizing on labor-arbitrage opportunities available in China and other low-cost countries. But in a future when the relative attractiveness of manufacturing locations changes quickly—along with the ability to produce large volumes economically—such standard approaches can leave companies dangerously exposed. That future, spurred by a rising tide of global uncertainty and business complexity, is coming sooner than many companies expect. Some of the challenges (turbulent trade and capital flows, for example) represent perennial supply chain worries turbocharged by the recent downturn. Yet other shifts, such as those associated with the developing world’s rising wealth and the emergence of credible suppliers from these markets, will have supply chain implications for decades to come. The bottom line for would-be architects of manufacturing and supply chain strategies is a greater risk of making key decisions that become uneconomic as a result of forces beyond your control.”

Anything that helps reduce complexity in the supply chain and improves decision making is a step in the right direction. The reason that supply chain segmentation is helpful is because it breaks supply chains down into smaller chunks. It’s like the old joke that asks, “How do you eat an elephant?” The answer, “One bite at a time.” Segmented supply chains are still pretty large bites, but they are easier to digest than one-size-fits-all supply chains.