There are a number of phrases one finds in articles and blogs about logistics. Two of the most common are “lean supply chains” and “flexible supply chains.” It seems to me that there is a built-in tension between these two concepts. When I hear the term “lean” I also think of the term “thin.” Consider a thin piece of wood, like a small strip of lath. A single piece of lath can be easily broken because it is thin and brittle. Stack a few pieces of lath together and they become much less brittle and much harder to break. My concern is that a lean supply chain can also be a brittle supply chain — and brittle supply chains can break. Does that mean that flexible supply chains are better than lean ones? Let’s go back to our example of the lath. If you stack too many pieces of lath together, the combined stack is stronger but it also loses flexibility. A rigid supply chain is the opposite of a flexible supply chain. There may be a better analogy for a supply chain than strips of lath, but I think it helps make a point. For supply chain executives, the challenge is to find the right balance between a supply chain that is too lean (i.e., is not robust enough but saves money) and one that is too rigid (i.e., is so robust and redundant that it is too costly to maintain and remain competitive). Let’s first discuss the lean supply chain.
Saurabh Goyal, a Consultant in Enterprise Solutions at Infosys Technologies, asserts that a lean supply chain will include minimal inventory on hand, delivery that is just-in-time, and a configuration-to-order strategy [“Lean supply chain – Is that always good,” Supply Chain Management, 30 March 2010]. You can tell from the title of Goyal’s post that he is not completely sold on the idea of lean supply chains. He continues:
“In most cases a manufacturer would want to have a lean system for the raw materials and the work in progress. But would [it] want this for finished products as well at the point of sale, I think not. Take the case of cell phones, with the lead time to introduce a newer phone getting reduced … manufacturers are wanting to keep minimal inventory levels in the supply chain and hence reduce losses due to dead sales for the product. This leads to a make-to-order scenario. But, does this hold good in today’s highly dynamic market condition. Today’s market requires the manufacturer to keep sufficient inventory at every Point of Sale and avoid missed business due to product non-availability. Even the lead time to introduce a competitive product with better features is decreasing by the day. As another example considering the Personal computer market, … consumers usually have similar requirements with standard configuration. So, … unless [a customer’s] requirements are very specific, [a customer will] go for a product that essentially delivers [the fastest] with lowest lead time. It is for this reason that we see ‘Fast track laptops’ from Dell and similar [offerings] from other vendors. Even in … consumer product groups like ‘Soap’, a customer [will] rarely wait for a [specific] soap to arrive on the shelf [instead of moving] to [more] a readily available brand. The concept of Lean does not hold true in these cases.”
What Goyal is saying in a nutshell is that, when just-in-time fails to deliver, consumers will find a more readily available product and manufacturers end up the losers. He continues:
“[Al]though a company wants to have a lean supply chain at the ‘goods in process’ leg, … at the same time it wants to have a ‘fat’ supply chain at the customers end. This means the business requires a hybrid system that caters to both push and pull. Pull in terms of having an agile system that responds quickly to any change in customer demands and Push in terms of maintaining sufficient inventory at the end of the system to be able to push sales by minimizing the losses due to non-availability of product.”
Goyal understands that maintaining a “fat” supply chain on the product side could “lead to a lot of dead inventory in the stock.” But he thinks that he has an answer that involves “agility, reverse logistics and modular engineering.” He explains:
“A high degree of agility would not only give more visibility and effective supply & demand management but also allow the management to keep track on the Inventory and the obsolescence of its products. The quick response across various points also leads to a higher customer relationship focus. … [However,] one still needs to take care of the inventory, which has a high degree of susceptibility to being obsolete. It is here that Reverse logistics comes into the picture. It would help in planning for the effective reuse, sale of the surplus and even disposal of these products. It will typically start from the Point of Sale and end at the point of manufacture/origin/disposal. This process would primarily help in clearing the shelf space for … new products, tend to minimize the losses due to no sale and also … create an effective system for reutilization of these products. To complement the process of Reverse logistics the manufacturers should also look into the aspect of Modular engineering. This means the product design [should] be as modular as possible leading to higher reusability of the components in the event of a product getting isolated. … A typical depiction of the model described above would be as follows:
I agree with Goyal that modularity, where it can be used, is a good idea. I’m not sure, however, that his model has solved the built-in tension between lean and flexible supply chains. Even Goyal admits some uncertainty by asking a series of questions: “Is this system viable? How would such a system be designed? What are the business complexities involved?” Noted supply chain analyst Lora Cecere calls herself a “lean bigot” [“Lean Bigot and Proud,” Supply Chain Shaman, 24 September 2010]. She explains why she “bristles” when she hears “the term LEAN in client interaction.” She asserts that “many companies have attempted LEAN programs with the best of intentions; but have moved backwards.” This backward movement, Cecere explains, results from “two disconnects.” The first disconnect results from a failure to appreciate fully all aspects of “lean” schemes. She writes:
“Muda needs Mura. Lean principles spring from supply chain excellence in the Japanese manufacturing industry. The approach of Toyota Production Systems (TPS), one of the most popular approaches, is to focus on the elimination of three types of waste: Muda (non-value added work), Muri (overburden) and Mura (unevenness). I find most companies focus on Muda, sometimes work on Muri, but seldom get to Mura. This is the root of my disconnect. I find that we have the supply chain team designing lean systems assuming that Mura will happen; while sales teams incented to induce more unevenness into the supply chain through sales incentives, price management, promotions, marketing, and trade deals. The issue is that 99% of companies have no demand visibility. They cannot see. Orders are not a good reflection of demand. So the supply-centered processes of source, make and deliver focus on lean, while the sales and marketing processes induce more variability. I find that it spins the organization out of control. What to do? Focus on the design of value.”
If you read Cecere’s posts over time, you find some common threads running through them. One thread is the importance of real-time data for supporting a true demand-driven supply chain. As she notes above, if you don’t have that kind of data you “cannot see.” That leads to her second disconnect: missing opportunities to eliminate waste because proper data is not being collected and disseminated. She writes:
“Greater value from Demand Driven. I have experienced many types of lean. So, while I believe in reducing Muda, I have also witnessed magic happen in many very SUCCESSFUL Kaizen events. I value the benefits from value stream mapping, continuous improvement and poka-yoke (error-proofing). So, you might ask, why does Lora bristle when she hears the word LEAN? Lean is frequently applied in manufacturing companies to reduce all the waste that we can see. But what if we cannot see? What if demand volatility is high, and sales processes have done anything but created Mura? In these cases, I believe that we must design for demand variability, improve demand sensing (the time to sense true channel demand), and shape demand based on market conditions. This is my second disconnect. Many clients confuse lean principles as being demand driven. Lean is an enabler of demand-driven; but can only be successful if you first build demand sensing and shaping capabilities FIRST that are market-driven. I find that most companies have sales-driven or marketing-driven programs that are anything BUT market-driven.”
For those unfamiliar with the word Kaizen, it is Japanese for “improvement” or “change for the better.” It refers to the philosophy or practices that focus upon continuous improvement of processes in manufacturing, engineering, supporting business processes, and management. [Wikipedia] Cecere concludes that if better collection and analysis of real-time data is not implemented in order to reduce the unevenness (Mura) in the supply chain, “we are not ready for lean. Yes, you may argue that we can apply lean on high volume predictable products. To this, I would agree. However, we cannot broad brush the concepts widely across the supply chain.” The editorial staff at Supply Chain Digest appears to agree with Cecere about the importance of data that drives supply chain decisions [“Buiding Sense and Respond Supply Chain Networks,” 17 June 2010]. The article provides an excerpt “from the groundbreaking report on integrated supply chain planning and execution written by SCDigest’s research organization Chief Supply Chain Officer (CSCO) Insights. … This excerpt provides a concise description of the third phase in the evolution to highly integrated planning and execution – moving to ‘Sense and Respond’ supply chain networks.” The article continues:
“Phases I (The Basics) and II (The Real-Time Supply Chain) take care of the fundamentals and move companies much closer to a state of integrated supply chain planning and execution. But getting to a true ‘sense and respond’ network, in which operational and even tactical planning blur with execution processes, is where the supply chain is ultimately headed – and faster than many may realize. What is close to available today and will become commonplace in the future is the ability to see in near-real time a complete picture of supply and demand data, across multiple levels – from supplier’s supplier to customer’s customer. The view will also not be ‘point-to-point’ but ‘many-to-many,’ requiring collaboration and synchronization around supply chain execution and tactical planning – in near real-time. But it is a lot more than just technology, of course – it is completion of the journey from a forecast-driven supply chain, in which companies build supply chains around trying to predict demand, to a customer-driven one that is organized to respond to demand with lightning speed.’ While the picture will look somewhat different from industry to industry, ‘build to stock’ will ultimately give way to ‘build or customize to demand’ in virtually every one of them. Long production runs and costly changeovers will give way to much more flexible factories that can rapidly switch gears to cost effectively make short runs of product based on the latest near real-time demand data.”
That article provides an excellent segue from discussions about lean supply chains to flexible supply chains. Dan Gilmore, Editor-in-Chief of Supply Chain Digest, breaks down supply chain flexibility into two types:
“Micro flexibility: … How fast a supply chain could detect and respond to issues and opportunities in the short term – maybe even right now. The truck is late, demand suddenly surges, a customer needs some sort of special packaging or handling: how fast and how effectively can this these changes and needs be managed?
Macro flexibility: … The speed at which a company’s supply chain adapt and execute new strategies and programs to support changes in overall company strategies or market place changes. Easy example: company decides it wants to build a consumer direct e-commerce channel, maybe with new SKUs in the mix. How fast can the supply chain out in place what it needs to do to make that happen? Another: Dell decides to move into retail.” [“What is Supply Chain Flexibility?” 4 March 2010]
Gilmore writes that he’s “not sure ‘micro’ and ‘macro’ were really the best words to choose, but hopefully the sense of them is clear.” In a subsequent article, Gilmore writes about “Defining and Measuring Supply Chain Flexibility” [23 July 2010].
“A couple of months ago, I wrote a column about supply chain flexibility, focused around the fact that most everyone seemed to agree that SCM flexibility was important, but how to define it and measure it was not clear at all. … But we still didn’t have much of an answer to a real definition – and certainly not close to how supply chain flexibility might actually be measured. So I decided to reach out to some supply chain friends and colleagues whom I thought might be able to add some clarity here. My questions: What is SCM flexibility, and can it be measured?”
Gilmore started with Dr. Lee, who responded:
“‘Flexibility should be measured with a time-element in it,’ Dr. Lee told me. ‘Hence, the flexibility to change volumes and mix must be time-based. Flexibility one month out versus flexibility three months out are very different. We should look at the whole time profile of flexibility in evaluating an operation – what is the flexibility over different time frames?'”
Gilmore agrees that “time is a critical dimension,” but he still feels that Dr. Lee didn’t define supply chain flexibility in a way that it could be measured. He next turned to “Tom Dadmun, once Supply Chain VP at network gear maker Adtran and now somewhere higher in the org chart, had this to add to the subject”:
“‘To me supply chain flexibility is about responsiveness to the customer. We can all debate whether we should call the process ‘Demand Chain’ or ‘Value Chain,’ but the key here is to have the Supply Chain respond to the customer’s demand,” Dadmun said. ‘To be truly flexible is to have a responsive process at every link in the chain that will quickly respond to the new input signal,’ Dadmun added. ‘Who’s in this control tower overseeing the entire value chain, even beyond the 4 walls of the enterprise? The Chief Supply Chain Officer of course. And his or her measuring stick of success is a hierarchical list of KPIs [key performance indicators], with alerts triggered by real time events identified by control limits on the BI Dashboard.'”
I agree with Dadmun that alerts triggered by specific KPIs must be part of a flexible supply chain. The earlier these alerts are provided to decision makers the more flexible the system becomes. The next expert to whom Gilmore turned was Bob Nardone, “a former VP of supply chain planning at Unilever/Best Foods and now running his own consulting firm (Supply Chain Guidance).”
“He notes that ‘Defining Supply Chain flexibility concisely is a challenge and one that probably has different nuances for different businesses.’ That said, he believes that ‘supply chain flexibility is the ability to quickly respond to changes in demand while achieving your customer service, cost, inventory and return on assets (ROA) objectives.’ That’s an important nuance – can you respond rapidly without breaking the bank? He adds that ‘I don’t think flexibility can be determined by one measurement but rather by a composite of metrics that many businesses currently measure,’ and that ‘A business should only invest in the level of supply chain flexibility (supply and distribution networks, systems, and people) needed to be competitive and grow.'”
Cecere might argue that in the future maximum flexibility WILL be required to be competitive. The Supply Chain Digest article on sense-and-respond supply chains also leads to that conclusion. Nardone, I believe, is arguing for a phased approach to flexibility, one that keeps up with competitors or stays slightly ahead of them without spending so much to achieve flexibility that it makes the company non-competitive. That is a tricky strategy to follow. The next expert to whom Gilmore turns is Dr. David Simchi-Levi of MIT, who “has done some great work lately that looks very specifically at supply chain flexibility.” Gilmore continues:
“He offers a specific definition of what supply chain flexibility means for manufacturers. If you have 5 manufacturing plants for a given business and each one can only produce one family of products, you have 1X flexibility. If each can make two families, you have 2X, etc. ‘Full flexibility,’ or 5X in this example, would mean every plant can make every product. The question then becomes what are the costs and benefits of achieving these different levels of flexibility? Those trade-offs can be specifically measured, he says. In subsequent comments to me, Dr. Simchi-Levi notes the connection of supply chain flexibility to the growing importance of supply chain risk management. ‘Little can be done after a disaster has occurred,’ Dr. Simchi-Levi says. ‘Companies therefore need to plan their supply chains so that they can better respond to both mega-disasters and mundane operational problems.’ He added that ‘flexibility into manufacturing, supply chain, and network strategies is essential if companies are to respond effectively to on-going change. Of course, the question is how to achieve flexibility and how much of it is required since flexibility does not come free.’ And I like this. Dr. Simchi-Levi says that he defines supply chain flexibility as ‘the ability to respond to change without increasing operational and supply chain costs and with little or no delay in response time.’ In this definition, he says, ‘change refers to change in demand volume and mix, commodity prices, labor costs, exchange rates, technology, equipment availability, market conditions, the production and logistics environment or any supply disruption.’ Now we are getting closer.”
Finally, Gilmore turns to his “colleague Gene Tyndall of Tompkins Associates.”
“Tyndall observes that ‘Supply chain managers [tend] to be cautious about flexibility. While customers may want it, financial, security, and risk controls may limit the ability to achieve it.’ He adds that ‘flexibility usually comes with an added cost, and this must be weighed against the value proposition for it. Supply chain managers strive for predictability and consistency, mostly for cost control reasons. When flexibility is introduced, so often are uncertainly and exceptions.'”
Gilmore concludes with a sense of uneasiness “that for all the emphasis on supply chain flexibility and agility these days, we have nothing even resembling a standard definition or set of metrics to use to track where a company stands or what progress it is making (though Dr. Simchi-Levi is getting close).” I don’t think that Cecere or Goyal would disagree with that. Until true sense-and-respond supply chains are implemented, there will continue to be tension between those advocating lean supply chains and those advocating flexible supply chains.