Written by 11:46 am analytics, Big data supply chains, Global Supply Chains, Market-Driven, market-driven value chains, Network of Networks

When We Don’t Have Answers, We Need to Work Quickly To Build Them

My email box is full of the “ain’t it awful blogs.” The storyline starts with a supply chain story of a shortage–semiconductors, rental cars, plastics–and ends with an easy answer. My take? Nothing about the current state of the supply chain is easy.

The burning issue of change is deep in boardroom discussions. Supply chain leaders have the ability to drive change like never before. These days, it seems that the supply chain makes headline news daily. It seems that no commodity is exempt. The story usually ends with a “supply chain expert” offering a point of view. And, voila! Amazingly, as by some fate, the remedy usually matches what the technology or services company is selling. Yawn.

My perspective is different. I believe that we do not know the answer to fix today’s supply chain. To drive progressive change, I think that we need to rethink supply chain basics and unlearn from the past while driving process innovation. Those that think that they have the answer should be shown the street.

The only thing clear to me is investing in the current state will not help us build better.

Let’s Start With a Discussion on Inventory

I remember my first implementation of Distribution Requirements Planning (DRP). A guy by the name of Gary trained me. I remember well when we got to the safety stock calculation asking him how we updated the optimization engine for network variability. The warehouse was bursting at the seams and the calculation did not seem quite right. His answer was, “Don’t worry, variability will not change the answer very much.” Gary might have been right then, but not now.

When we look at table 1, we see burgeoning inventories in some industries and extreme shortages in others. The end result? Insufficient warehouse capacity and a lack of containers while supply chain shortages make headline news. How might you ask, “Can we have shortages in warehouse space while we also have empty shelves?” The answer? We are not good at inventory management. Companies are not good at planning. We cannot effectively use market signals and adapt supply chain processes.

Inventory Levels Over the Quarters of the Pandemic

Four industries–B2B Technology, Aerospace and Defense, Retail Apparel, and Food Manufacturing–grew inventory during the pandemic. In contrast, other sectors reached unprecedented shortages, including Semiconductors, Pharmaceuticals, Chemicals, Contract Manufacturing, and Retail Home Improvement. Look at the 72 ships off the coast of Long Beach (up 65% from a year ago). Does anyone want to wager with me how many of those ships have inventory that is not really needed in the supply chain?

Could Outside-in Thinking Help?

We started the pandemic with an average of twenty-one more days of inventory on average across industries than at the start of the 2007 recession.

Companies struggle to have the right inventories. The reason? Inside-out traditional and functional processes cannot sense and adapt to market shifts. Only 9% of companies actively design their supply chains and only 12% actively model the form and function of inventory. Most companies manage inventories by focusing on the establishment of safety stock models in DRP. This is not sufficient. The assumption is the reliability of the first mile and constant lead times. Today, neither is constant or reliable. If it was not so painful, the blind use of these assumptions would be comical.

Table 1. Definition of Form and Function of Inventory. Each of These Buffer Strategies Require Design and Bi-Directional Orchestration

While some believe that the current set of problems are based on things like JIT inventory or VMI, this is just not true. The schism is much more systemic and rooted into traditional process thinking. Let’s take an example. The unloading delays at ports increase in-transit inventories and the delays in unloading should increase safety stock buffers, but for most, they will not. Our current systems cannot drive this type of response. The gaps in inventory planning–what is needed and what is used–are huge. The requirements definitions are not in the current definitions of planning.

Figure 2. Comparison of Inventories Across the Years

Over the last decade, as we automated functions to make them more efficient, we focused on cost. Less than 2% of companies operate by a balanced scorecard across functions. Only 29% can see the total cost and less than 1% can actively track margins and customer service. So, as you read the headline news, don’t you think that it is time for us to rethink the basics of what we are doing? In many ways, the hardworking employees in the supply chain are working hard to eradicate the issues of traditional thinking. Don’t you think it is time to build better?

Building Outside-in Processes

For two decades, I have spoken on the need to build outside-in processes. Today’s supply chain processes respond, but they cannot sense. Data surrounds the supply chain, but it cannot be used because it is disparate. The reliance on relational database architectures and functional excellence make much of what we have today in technology legacy.

No company that I work with knows how to use point of sale data to mitigate the bullwhip impact and decrease demand latency. Over 80% of the implementations of demand sensing are not used. Frustrated, I started Project Zebra.

The goal is to define and test process canonicals. Over the past year, the project, working with business leaders defined and mapped processes to use channel and supplier data to drive bi-directional orchestration. (Bi-directional orchestration is the constant redesign of the supply chain across make, source, and deliver to improve outcomes and reliability.) Current processes single-thread tactical planning across functions and the data models for deliver (DRP), make (MPS), source (MRP or DDMRP), or transport (TMS) do not enable orchestration, or process interoperability. The goal is to redefine the models to drive interoperability across time horizons.

Today, as headline news reaches epic levels, we are at an impasse. Technologists respond to RFPs based on yesterday’s technologies and business leaders are hamstrung to invest only when there is a clear ROI. IT leaders blindly follow the path of IT standardization. The industry is not good at process innovation even though it should be clear to all that today’s processes are not equal to solving the issues of the headline news stories.

In Project Zebra, we are currently working with BSH (manufacturer of home appliances) and Western Digital to test two of the eighteen capabilities identified. The process map from the last discussion is share below. The largest hurdle for the process development is “unlearning” by senior supply chain leaders–both senior technologists or business leaders. In the past, we have tried to drive change management without aligning teams through unlearning processes. To learn more about the methodology and processes and the potential impact on results join us for the October 14th event and sign up your team for the three follow-up zebra training sessions.

The second phase of Project Zebra will start in January with five companies that want to engage in deeper future development. The outcomes of the project will be built in the future technology roadmap of the SCOR model for planning. I hope to see you there!