Wednesday, June 27, 2012

Demand Forecasting—Why Collaboration Is Key

Forecasting demand is critical to efficient and effective inventory management. Unfortunately, it’s not easy. We now have intelligent and efficient automated systems to help provide forecasts, but they can’t handle all exceptions or predict external factors—and we have certainly experienced a few external “factors” in the past several years (2008 financial crisis and Japanese earthquake, to name a couple).

Such events over the past few years have provided some key learnings and many OEMs are doing pretty well with forecast calculations. However, not all are fully communicating this information. While about 95% provide forecasts to external suppliers, only about 75% provide forecasts to internal suppliers. Why isn’t there more internal collaboration? Given the complex and often unpredictable nature of demand forecasting, communication and collaboration is vital to success.

At the 2012 NAPB in Chicago, we discussed some recent forecasting & inventory management success stories. What they all have in common is improved practices of collaboration and communication. Below are the key takeaways:
  • Get all responsible and accountable parties on the same page. Forecast metrics often require the input and oversight of several different parties. However, it’s rare that everyone involved in the compilation of this information has the same goal, involvement, and stake. Without such common ground, components of the metric can be disjointed or misunderstood. Aligning the understanding of the metric’s definition, calculation, value, impact, and use ensures that anyone responsible or accountable for the forecast (in however small a capacity) is well-informed and aware. And that means a better, cleaner forecast.
  • Meet face-to-face. So, how do you get everyone on the same page? Email? Not great. Phone? OK, but we can do better. Email and phone do little to open up communication and facilitate discussion. They’re fine day-to-day and in a pinch, but they don’t compare to meeting in-person. Being together as a team to discuss calculations and processes reinforces each team member’s responsibility and helps to clarify points of confusion and address concerns.
  • Share information consistently and often. Forecasts should be shared amongst all involved parties and should be shared with some frequency. One OEM with recent success developed a process of communicating forecasts and planned shipments on a monthly and weekly basis. This OEM provides reports at the company and the supplier level with sales, receipts, and inventory forecasts by month. With these monthly forecasts in mind, weekly shipment and order schedules are created for each warehouse based on individual capacity. These consistent reports keep suppliers and warehouses prepared and in the loop. This transparent and methodical process of information sharing has led to great process and efficiency improvements.
  • Address cultural differences and barriers. Often, parties involved in the calculation and dissemination of demand forecasts are scattered across the country and the world (just one reason why it is so important to meet face-to-face, at least occasionally). Team members may have very different cultural values and work styles. One OEM recently had issues coordinating between teams in the US and Asia. In addition to metric and process improvements, getting “back to basics” and reinforcing many things that had been miscommunicated, lost in translation, or falsely assumed, led to a more cohesive team environment, which paved the way for more efficient and innovative work.
  • Shake things up—don’t stick with the status quo. Innovative thinking is typically at the core of real success. One OEM showed us how powerful this can be. This OEM had a decent forecasting system, but it was a bit cumbersome and identified many exceptions, all of which had to be manually reviewed. So, the OEM developed a secondary, offline system through which to run all such exceptions. This generated a better adjusted forecast and reduced exceptions needing manual review. It required an intensive set up process, but collaborating with a diverse team rapid algorithm development and testing led to a more accurate forecasting system.
Bottom line: Forecasting demand is a complicated process with great impacts on inventory, fill rates, and ultimately customer satisfaction. While a lot of work goes into these forecasts, the importance of team dynamics and information sharing is occasionally overlooked or undervalued. Communication and collaboration in the forecasting process is important. This year’s success stories demonstrate great improvements in this area. Those who aren’t fully collaborating internally and externally need to get on board.

No comments: