By making the flow in the supply chain at the global level even and uniform to the most practical extent, the designers of the supply chain are able to detect systematic variations quickly. This detection is based on managing random variations using well-designed systems and processes that adjust to the random variations with small and permissible perturbations. Sales and Operations Planning and Production Scheduling and Operations revealed how adjustments are made continuously as more accurate information becomes available. Possible adjustments are prioritized, with volume adjustments being the most difficult to make. Volume adjustments require overall consensus, whereas parts that are supplied from nearby sources can be adjusted within preset limits. So, if the mechanisms that have been prescribed to address variations fail to match demand and supply, then it is a signal that a systematic shift has occurred. The dealer interview describes how data regarding product volumes and appropriate sales stimulation are continuously shared between Toyota and dealers. The inbound logistics process describes how milk runs ensure pickup in small volumes from suppliers on a continuous basis. All of these processes ensure a flow view of the entire supply chain.
Managing Visibility Using a Pull System
The pull system is used to link successive production units together. Clearly, in a supply chain the role of a pull system is to match demand and supply by linking the flow of information to the flow of material. The extraordinary simplicity of having suppliers that are located in close proximity to the plant minimizes the need for coordinating activities at the operational level (but not at the planning level). The pull system is also Toyota’s way of executing so that variations become visible. Let us take for example the famous seat supply case,10 in which a supplier delivers seats to the assembly line. As cars leave the paint area, a signal is sent to the seat supplier that indicates the seat required. Seats are made in the order in which the signal is received and delivered by truck to the assembly line just-in-time for the seats to be put in place at the final assembly lines. In the normal course of things, a few seats might be damaged and a few might not fit the car. However, the case describes a situation in which not only do several cars wait in the “clinic area” following assembly for replacement seats but also no one seems to be aware of the seat problem. Has the pull system failed? One view is that the problems accumulate because of the pull system. Another, and our view, is that the pull system worked. It has revealed systematic problems that need to be addressed. One or more seats waiting is to be expected because of random variations. Several cars waiting for replacement seats is a sign that something has gone wrong.
But how does the supply chain benefit by forcing the seat supplier or a tire and wheel assembly supplier to produce with a pull system? It forces the supplier to standardize the process, to level production, and to make its own suppliers deliver in a smooth fashion. Thus, if a critical part holds up production, the Toyota engineers as well as the seat supplier become alerted to the problem early. The benefit of managing random variations using the capabilities built into the systems is that it allows for an early warning of truly systematic problems. Moreover, it brings to bear the expertise of the supply chain community to solve problems.
Managing Visibility Using Visual Controls
Visual controls have several purposes. The first is to make problems visible, thus enabling the team member to recognize variation and forcing a decision to be made within a given time. If the decision cannot be made by the team member, the case is deemed to be a special one and should be escalated. In a supply chain, as we have seen, it includes setting limits on the mix variations allowed and using visits to suppliers and dealers to understand their processes. Toyota Motor Corp. Chairman Fujio Cho recently extended the definition of going to see the problem to, “Have you seen it yourself?”11 We have also described the practice of sending engineers to visit suppliers and dealerships. That customary action reveals the second purpose of visual controls, especially in a supply chain: to learn about different systems firsthand. One might wonder how much firsthand learning is necessary to manage supply chains. We shall address this issue in the section on The Theory of Learning. A third purpose of visual controls is to pro- vide transparency to any observer regardless of function or ownership. The Toyota Way document states: “We share a common understanding of actual conditions by using visual controls.”