Steve St. Angelo, president, Toyota Motor Manufacturing Kentucky (a.k.a., the Georgetown Plant, where Toyota produces Camrys, Camry Hybrids, Solaras, Avalons, engines, and plastic parts) and senior vice president, Toyota Motor Engineering & Manufacturing North America, recalls working at NUMMI (New United Motor Manufacturing, the General Motors-Toyota joint venture plant in Freemont, California) back in 1995*. Actually, he was there not to be working at NUMMI per se, but, as a GM executive advisor, to be there learning the Toyota Production System. At the time, Gary Convis was the general manager of Manufacturing at the plant. St. Angelo says that he was rather insistent that he do something rather than just observe. So Convis said that St. Angelo could get involved, could do something. And that something happened to be working the line for 12 weeks. Working every job, rotating through the teams. And he would have to work both shifts. St. Angelo thinks that maybe Convis was trying to dissuade him. He wasn't dissuaded. He worked through the jobs. He says that some people didn't know that he was from Detroit, from General Motors. They thought he was just another new hire. He learned the jobs. Learned the Toyota Production System in a way that reading and observing just wouldn't have made possible. He was the group leader in an area where there was a particular problem in terms of output. He says that the team worked together and went from 100 defects per day to none.
He learned his lessons well. He cycled out of NUMMI but returned in 2001, when he was named president of the operation.
Problems are part and parcel of the Toyota Production System. And in the manufacturing plants, one of the ways to draw attention to a problem is to pull the andon cord. St. Angelo says that he'd read so much about TPS and about andon that he imagined that he was, in effect, a "sensi of andon." But then he was faced with having to pull it. He thought that everyone was looking at him. Staring at him like he'd done something wrong. The line was stopped. He did it. It was a very different feeling.
Realize that St. Angelo was someone from the GM system. From a place where, at that time, andon cords were as common as Barcaloungers. He describes a basic difference between the way work was done: Toyota is a process-driven company; others are result-driven. By which he means results at the end of the day. If you're worried about part count, then you don't pull an andon cord. "Andon exposes problems," he says.
Fast forward to today. When the Georgetown Plant is prepped and ready for the production of the Venza crossover. Dimensionally bigger than the Camry upon which it is based, they found that it was necessary to change about 75% of the processes in the paint shop. But in final assembly, it was just 4%.
A key to the difference, he says, is "SPS" or, "set-part strategy." Which is an approach to kitting. He says that in the past, the kitting was performed such that all of the parts for the variants of vehicles (those, say, needing light interiors and those needing dark, which means a vast array of different trim pieces, grommets, and the like) were presented to the operator on a roller rack. As the cars came down the line, the operator then made a determination of what was required. But with the new approach, SPS, the parts that are presented to the operator are only those necessary for the vehicle. There is no reading of a manifest or scanning a barcode. The parts are there, ready for installation.
In some cases, there are no racks along the line. In some cases there are AGVs that follow a car along the line, carrying the kitted parts for the vehicles being produced.
Which leads St. Angelo to make an analogy about what should be on the line-and what shouldn't. He likens the main assembly line to "ocean front property." He explains that generally, the property is expensive and so if you're going to build a house on the property, you probably build an expensive dwelling. He contrasts this with farmland, where the cost per square foot is significantly less. This is the area where the SPS operations are performed, where the real estate is not so expensive.
Another statistic related to the Venza that he says some of his industry colleagues find to be vexing is that there were 40% fewer engineering changes. If that's not enough, during trial production, defects per vehicle decreased by 27% compared to previous new model trial builds-and realize that the vehicle that launched prior to the Venza in Georgetown is the Camry. These are consequences of identifying problems up front. And of what he calls "built-in quality with ownership." He says that the way they're working within Toyota-be it at the plant or at the Toyota Technical Center in Ann Arbor, is that the number of people who are assigned to the program is comparatively large at the start. Everyone is encouraged to find and fix problems. He draws a pair of intersecting lines forming a wide X-shape. He explains that in traditional product development practice, the number of people assigned to a program grows as the vehicle gets closer to production. In this approach, chances are problems are going to be found closer to start of production than earlier, if for no other reason than that there are more people working on it. In the Toyota approach, it is pretty much the opposite. Early discovery of problems lead to fewer late production changes and improved defects in production.
*St. Angelo spent more than 30 years at GM. He started at the Fisher Body Fleetwood plant, and worked at many operations including the Lordstown, Willow Run, Oshawa, and Lansing plants. Prior to retiring, he was manufacturing director of General Motors de Mexico. He joined Toyota in 2005.