Operations managers typically use applied mathematics and sophisticated algorithms to perform complex scheduling of, for example, a fleet of drivers and trucks to optimize resources and minimize costs. Operations management relies on computer-generated models to take into account all facets of a complex business operation. But models haven't been good at capturing the customer side of the business equation, says William Pierskalla, a professor of operations research at the University of California at Los Angeles, speaking this week at the Institute for Operations Research and Management Sciences in Montreal. "The customer is a key part of any business solution, and the more you can integrate those factors into your business models, the more successful your planning will be," he says.
Warren Powell, director of Princeton University's operations research department, concurred, noting that "humans tend to know things that computers don't."
Praxair Technology Inc., a maker of industrial gases, uses operations management to improve the timely delivery of products to customers. "We had a sense of urgency about increasing our business efficiency. We wanted to not only reliably deliver products to customers such as hospitals but to do so in a better way than we had been doing," says Laurie Dutton, logistics manager for Praxair in Tonawanda, N.Y.
Dutton and her team worked on the problem for a decade, delivering various versions of a routing and scheduling application to the company's truck dispatchers, who would use the tool for a while, then abandon it. "We keep polishing the model, but ultimately we were polishing dirt because the system design didn't take into account dispatcher feedback." It wasn't until Dutton moved her desk into the middle of the dispatch center and listened to how these internal customers worked did the system deliver what dispatchers and their customers needed.