Supply Chain Agility: If you know it when you see it, do you need to define it?
In a recent exchange with Lora Cecere on Twitter in which I stated that “agility” is a requirement for any modern supply chain, Lora challenged my statement by stating that there isn’t a standard definition for agility. More precisely, Lora’s point was that there isn’t a standard way of measuring agility, which of course relies on a standard definition. OK, I get Lora’s point, but, like that famous statement by a U.S. Supreme Court judge when asked to define pornography, he replied that “I know it when I see it.” So let us explore some of the terms, particularly agility, used to describe today’s supply chains.
The many terms and definitions…are we all saying the same thing?
As long ago as 2004, Hau Lee used “Agile” as one of the cornerstones of his Triple-A Supply Chains, the other two being “Adaptable” and “Aligned.” Hau’s full definition is:
Agile: They respond quickly to sudden changes in supply or demand. They handle unexpected external disruptions smoothly and cost-efficiently. And they recover promptly from shocks such as natural disasters, epidemics, and computer viruses.
Adaptable: They evolve over time as economic progress, political shifts, demographic trends, and technological advances reshape markets.
Aligned: They align the interests of all participating firms in the supply chain with their own. As each player maximizes its own interests, it optimizes the chain’s performance as well.
… 70% of executives who responded to a recent Accenture survey expressed concern about their inability to predict future performance, and more than 80% worried about the overall resilience of their supply chains in the face of unrelenting market challenges.
This is a great summary of the market drivers that are forcing manufacturers to adopt the concepts behind dynamic supply chains. But what are these concepts? From Mark’s description of the market drivers, we now we have three terms: Dynamic, Agile, Resilient. Mark does not define these terms, although he adds other terms including Flexibility, Adaptability, and Responsiveness. Nevertheless, Mark does a great job of describing the primary characteristics of a dynamic supply chain:
- “Portfolio of supply chains”
- Dynamic operating model
- Sense, Shape and Respond
- Flexible Product Strategy
- Agile Execution
- Operational Hedging
For me, all of these terms come down to agility in the face of demand and supply volatility, the key point from the Accenture survey quoted by me above. The best definition of agility comes from Martin Christopher at the Cranfield School of Management in an article titled “The Agile Supply Chain: Competing in Volatile Markets”:
Agility is a business-wide capability that embraces organisational structures, information systems, logistics processes and, in particular, mindsets. A key characteristic of an agile organisation is flexibility. Indeed the origins of agility as a business concept lie in flexible manufacturing systems (FMS). Initially it was thought that the route to manufacturing flexibility was through automation to enable rapid change (i.e. reduced set-up times) and thus a greater responsiveness to changes in product mix or volume. Later this idea of manufacturing flexibility was extended into the wider business context (3) and the concept of agility as an organisational orientation was born.
Agility should not be confused with Lean. Lean is about doing more with less. The term is often used in connection with lean manufacturing to imply a zero inventory, just-in-time approach. Paradoxically, many companies that have adopted lean manufacturing as a business practice are anything but agile in their supply chain. The car industry in many ways illustrates this conundrum. The origins of lean manufacturing can be traced to the Toyota Production System (TPS), with its focus on the reduction and elimination of waste.
What I love about this definition by Martin is the comparison and contrast between agility and lean. This contrast is brought out in a blog titled “Supply Chain Concept,” in which the author states that the difference between agile and lean supply chains is as follows:
- Forecast at generic level
- Economic batch quantities
- Maximize efficiencies
- Demand driven
- Localized configuration
- Maximize effectiveness
What is not stated in the diagram above is that if an equivalent diagram were drawn with margin on the Y axis, it is likely that the 20% of the products that make up 80% of the total demand would only account for 50% or less of the profit. This is because many of the products in the “agile” space have higher margin, often because they are new products.
In a blog I wrote some time back titled “One size does not fit all,” I referred to some great work done by Matt Davis of Gartner (“Frameworks to Design and Enable Supply Chain Segmentation” Davis, M., Gartner, Inc., 19 May 2011), that reinforces this point. Matt points out, in the the chart below, the need to differentiate between “agile response” and “efficient response” at different stages of a product’s life span. I noticed his inclusion of the word “response” when reading the article for the first time because it implies response to volatility. And of course, there is the overlay of the profitability curve emphasizing when it is best to deploy an efficient (lean) rather than an agile response to satisfy customer demand profitably.
Staying with Gartner for a while, they have been promoting the concept of Demand-Driven Value Networks (DDVN) for many years. A key aspect of DDVN is Demand Translation. In an article titled “Supply Chain Strategy for Industrial Manufacturers: The Handbook for Becoming Demand Driven” (Barrett J., Barger R., Gartner Inc., 2 September 2010), which includes the diagram below, Jane Barrett and Ray Barger state that:
The ability to accept or decline orders based on profitability and to consistently align to business strategies is difficult to cultivate, but it’s a characteristic of mature, Stage 4 demand-driven leaders.
Orchestration requires the ability to translate complex go-to-market strategies, new offerings and supply networks into plans to meet business goals most profitably. Like cogs in a transmission, the demand, supply and product functions within an organization must be aligned and synchronized for this to happen. Each function must forecast and plan based on its own strategy, while the cross-functional S&OP or a similar process brings them together to orchestrate the demand-driven response.
As you can see from the diagram above, demand translation is at the very heart of a DDVN, and, as Jane and Ray state, it is a characteristic of mature demand-driven leaders. To me, this is the very essence of a dynamic supply chain.
If you can’t define it, can you measure it?
Let me refer back to Lora’s challenge. I admit that, beyond characteristics described by Gartner, Cranfield, and Accenture, amongst others, I am not sure that I can define a dynamic or agile supply chain. But as sure as heck I know it when I see it, and I think we all do. But what about a standard way in which to measure agility/flexibility? To be honest, there isn’t even a standard way in which to measure forecast accuracy. Many people use MAPE (mean absolute percent error) while others use a weighted MAPE, with the weighting factor varying between revenue, units, and margin. Even simple MAPE caused a heated discussion on LinkedIn where some people prefer |Forecast – Actual|/Forecast and others prefer |Forecast – Actual|/Actual, which is the APICS definition. While there isn’t a consensus on how to measure something as simple as forecast error, I am quite relaxed about not having a precise measure of something more complex like agility.
Nevertheless, Lora has a point, because if you can’t measure it you can’t control it. As a starting point, I would like to suggest that there isn’t one key performance indicator (KPI) that can be used to measure agility, but that the following list, in descending order of importance, is a good start:
- On-time delivery (OTD) to original customer request date and at standard cost
- All too often we measure OTD to promise date
- It is easy to expedite materials to meet a customer request, but doing so profitably is a true measure of agility
- Most importantly, this KPI is consistent with the Gartner definition of Demand Translation
- Inquiry-to-quote lead time
- The time it takes a company to respond to a customer request with a promise date for either a new order or a change in an existing order
- I’d like to suggest that this should be less than 10% of the order-to-delivery lead time
- This measures the efficiency or speed of converting cash to supplies and then into products that can be sold
- Some people object to C2C because it also measures Days of Sales Outstanding (DSO), which is essentially the tie from when the customer gets an invoice to when they pay, and Days of Payables Outstanding (DPO), which is essentially the time between when materials are receive to when a supplier is paid. People object because DSO and DPO are not under the direct control of Supply Chain. This is true, but many companies try to hide channel inventory at dealers and distribution partners in DSO, and raw material inventory liability as DPO.
In closing, Mark Pearson makes this statement about transitioning to a dynamic supply chain:
Moving from the “integrated” to “dynamic” supply chain model enables companies to view their supply chains as adaptable ecosystems of processes, people, capital assets, technology and data. They strive for flexibility where it matters and focus their efforts on operational agility that drives profits, and not just short-term efficiencies
Sounds like an Enterprise Control Tower to me, especially when you consider that Stage 4 of the Gartner DDVN maturity model is called “Orchestration.”
- I am adamant that an accurate forecast does not reduce demand volatility
- A must-attend IW webcast with Cisco VP, Karl Braitberg and demand management researcher, Dr. Larry Lapide
- SCM predictions for 2010: Lessons from the retail world
- Improved Planning: Looking for clearer forecasts as recovery nears
- Envisioning the new normal and other supply chain phenomena