Should I implement a kanban "point solution"?

[Randy Littleson]

Just the other day I became interested in what had evolved regarding kanban tools and I did a quick search of the internet. A couple of hours later, I was more than a bit surprised by the number of point solutions that were targeted at helping companies identify where kanban might be utilized to drive significant savings in inventory. Some companies even offered 30 day trials and promoted easy integration approaches for acquiring the data required to deliver value. Given my past experience in implementing Lean, I was intrigued by the sales pitch to implement a point solution rather than leverage kanban as part of a much larger initiative to create a Lean value chain. I’m not at all convinced that the uninitiated will recognize the very importance difference and potential limitations or pitfalls of a purely point solution.


What these companies are offering is a tool that will essentially do the math to recommend kanban parameters (total kanban inventory and re-order point quantities) and incorporate logic to address demand variation as a safety factor. Based on the results, the suggested level of inventory is compared to the actual levels and inventory savings is calculated. Seems simple, and to a large degree it is, but the decision is completely removed from the value chain perspective. At this point you are probably thinking that I am offering some kind of zen based perspective driven by a Lean ideology. The truth is much simpler, the source of data for these solutions is historical stock issues (for demand variation) and MRP calculated demand for the future requirements. This information can be the equivalent of a mile distant from the actual customer demand associated with those parts. Between the customer demand for the end item and the component demand can be a wide of variety of “Just-in-Case and Efficiency Based” planning strategies at various subassembly levels including lot sizing, minimum and multiple order quantities, and various safety stock strategies. Each of these strategies contributes to lumpy demand patterns at the component levels and a correspondingly erroneous picture of true demand requirements and variation. While implementing a kanban system at the component level alone without consideration for the larger value stream “might” yield some inventory savings, it is little like putting in a 10 watt bulb in your clothes closet. The light isn’t bright enough to dress by and the result will likely have some undesirable and comical side effects.



Taking the value stream approach means that you look at all of the supply chain connections from the customer down to the shipping dock and the rules which govern the execution process. It is holistic in nature and links how the decisions at each stage influence later stages. Demand at all levels is linked to the customer expectations and the goal is to eliminate unnecessary waste at every stage. Inventory buffers are strategically determined based on manufacturing cycle times and customer responsiveness expectations. For example, rather than maintaining inventory to support demand variation at the component level, this inventory maybe maintained at a higher assembly level to support a responsiveness expectation that cannot be met otherwise. In addition, the Lean efforts would seek to minimize setups and the need for batch sizes with the ultimate goal of single piece production (not always practical or possible where investments to achieve this are unrealistic). The bottom line is that the holistic view makes sure that the investment in kanban systems is maximized and true aligned with ultimate goals for minimizing waste while maximizing the value delivered to the customer. Therefore, beware that investing in the implementation of a kanban system without evaluating the connection to the rest of the value stream may not yield the benefits expected. In fact, the kanban inventory levels that are calculated may be significantly more than is really necessary and the desired level of flexibility may still not be adequate to meet customer expectations based on the cycle time distance between the component inventory and the end product.


Kerry Zuber is the Director of Business Consulting for Kinaxis, provider of the on-demand RapidResponse service that empowers multi-enterprise manufacturers with the collaborative and integrated demand-supply planning, monitoring, and response capabilities.