Kurt Zetah with Dow Chemical provided this presentation. Kurt is Network Optimization leader and Mergers and Acquisitions leader for Dow's Supply Chain. Dow has 200 manufacturing sites in 38 countries, with $54 billion in sales and 46,000 employees, serving customers in 160 countries.
Dow began supply chain modeling in the 1980s, using linear modes with regard to feedstock selection and, to a limited extent, customer sourcing. They moved into production planning and distribution center planning in the 1990s using statistical modeling; into sophisticated sourcing analysis in the 2000s with mixed integer modeling; and into stochastic models today. Back in the 1980s and 1990s, the data collection was largely manual, but as they have become more integrated from a systems perspective, they have been able to reduce modeling time from months to days or even hours now. But they want to become even faster to be able to respond to rapid changes in raw material, transportation and other costs. This is in recognition of the fact that supply chain costs have become increasingly important – and increasingly recognized as important within the company.
In summary, Kurt noted that modeling is a requirement to compete today – "If you're not doing modeling, sell your stock in your company." You need to be able to model faster and more accurately, and you have to institutionalize the process to ensure success over time.