Remembering that too much flexibility undermines profitability and requires heroic efforts to provide expected service.
I recently visited a facility that was a paragon of flexibility. The team met almost every day to review their schedule, honor last minute customer requests and ensure that all of their production assets were running at or even slightly above capacity.
Thinking strategically, the company had recently added some new assets, processes and capabilities. Each new addition created even more opportunity for new customers and greater volume. New technology meant more configurations, options for batch sizes, and entirely new products. The increased capabilities made forecasting more complex, but the team diligently persevered and relied on time honored forecasting approaches, which had always worked in the past.
They consistently tried to follow leading practices including Managing-by-Exception and addressing the highest priority challenges each and every day. However, due to some glitches and unavoidable disruptions the facility had fallen behind schedule and was trying to catch up with demand. All this was leading to eroding profitability and missed customer commitments. Batch sizes were trending down and with them unit costs were trending up. Standard cost assumptions that had previously been effective estimates were turning out to be wrong in the new more flexible paradigm and the scheduling team was drowning in last minute changes. The entire operation was in constant firefighting mode and a culture of heroism had developed.
We’ve seen this pattern before – the constant and pervasive pressures to reduce costs, lean inventory and increase product portfolio while offering an ever more responsive, adaptable and immediate customer service is greater than ever. In order to compete, companies in every industry feel the need to offer more SKUs, more configurations and shorter lead times while shaving ten, twenty or fifty percent off their customer facing lead times. These pressures collide in the manufacturing and distribution stage of the supply chain where customer service demands increasingly smaller batch sizes with decreasing lead-time to best fulfill a customer that does not want to buy or hold months of supply for a product.
But what is the solution? Should we simply acknowledge that the new demand patterns are fickle and embark on a quest to reduce changeover costs to zero and minimum batch sizes to one? Is forecasting so impossible that we should simply throw our hands up and spend our time and money building flexibility to react to any eventuality? Before we surrender to uncertainty we must first endeavor to better understand it.
End-to-end supply chain modeling and clear customer expectations can be very helpful. For example, manufacturing, particularly in “commodity” industries tends to treat customer demand and delivery dates as if immutable and inscrutable. Moreover, in the name of customer service, the culture would rather rely on heroics than push back on the customer. First and foremost, a set of clearly defined parameters regarding the key elements of customer orders and committed lead times based on key order parameters are absolutely essential. Arbitrary or vaguely estimated lead times are often either too long or too frequently missed.
This is where a solid Sales, Inventory and Operations Planning (SIOP) comes into play. Establishing such a disciplined process enables a good lead-time commitment process. Both processes rely on a solid model of the production facility in question. Equipment capacity, up time, changeover costs, competing capacity demands, etc. should all be captured and modeled. This allows for programmatic and optimized sequencing and slotting of orders, which is a key part of the near term “Operations Planning” in SIOP. The model also allows for effective projections of capacity allocation into the future. Armed with this, the plant is able to issue solid commit dates and consistently hit them. In the end, this predictability is better for the plant and customer than making more ambitious projections and then missing them five to ten percent of the time.
In short responding to uncertain and unpredictable demand by better understanding constraints through modeling, improving discipline in planning and setting clear and consistently achievable expectations with customers is better than allowing your operation to be perceived as unreliable by customers and unprofitable by management.
Read one of our case studies on end-to-end optimization below or click to watch Dr. David Simchi-Levi discuss SIOP in this short video.
Written by Jay Watt, Director of Analytics for OPS Rules Management Consultants