In the paint booth, it was accepted that “first-time-through” quality would not be very high due to contamination which was very difficult to eliminate, but that skilled paint specialists could eventually bring the body paint up to an acceptable level. Finally, once the moving track was installed in 1977, the operating philosophy was to quickly put all of the parts on the car, then test them as a system after the car rolled off the line, and to rectify errors in a highly skilled troubleshooting and rework process which eventually produced a product with a world-class low level of defects as reported by customers. Skilled work was, therefore, defined as the ability to operate specific machines and to diagnose anomalous conditions during long work cycles and to take corrective action on a case-by-case basis.
This approach was also applied in downstream portions of the product development system, where manufacturing engineers took product designs and either figured out how to make them or secretly reengineered them. Even worse, as anyone owning a Porsche has learned, there was practically no attention to serviceability because the voice of the service bay was simply not represented in the system. In consequence, a whole new skilled trade was created around the world, the Porsche mechanic.
The Porsche craft tradition had great appeal to many workers because of the long cycles and the opportunity to put every worker’s considerable skills to the test continuously. It also appealed to many managers because there was no need to take up the messy and unpleasant task of confronting the cause of problems at upstream stages and rectifying them at the source.
The Crisis
Porsches offered truly superlative performance based on a deep technology base and filled a special niche in the market for true sports cars just tame enough for everyday use. As a result, it was difficult for either giant car companies or tiny specialists to challenge Porsche. Sales volumes were too low for the high-volume car companies to bother with, reaching only 33,000 units in the peak year for the highest-volume model, the 944, and never exceeding 21,000 for the upmarket 911. Smaller specialist firms who might have copied Porsche’s product philosophy and worked cost-effectively at low volumes lacked the necessary product technologies built up over many years by Porsche’s consultant engineers.
However, the firm’s special situation also created vulnerabilities. For one, any model change was truly a “bet-the-company” proposition, so over time the management erred on the side of caution. The 928 model was planned as a replacement for the 911, but when customers balked at the 928’s front-engine, rear-drive design, the 911 was simply continued indefinitely alongside the 928. Another critical vulnerability was that a majority of those with the money and desire to buy a Porsche in the 1980s lived in North America, while practically 100 percent of Porsche’s value was created in or near Stuttgart.
As a consequence of these vulnerabilities, the boom year of 1986, when Porsche sold a record 50,000 cars (62 percent of them in North America), gave way to nightmare years from 1987 on as the mark strengthened against the dollar and sales steadily tumbled. By 1992, Porsche was selling only 14,000 cars worldwide and only 4,000 rather than 30,000 in North America. (Table 9.1 provides a production history of the Porsche Company.)
The initial response of the Porsche and Piech families to the sales collapse was hesitation. They hoped it was only a blip in the market. However, by 1989, the downturn was continuing and the family brought in new senior management with a marketing focus to revitalize sales. Arno Bohn, the marketing director of the Nixdorf computer company, was hired as the new chairman to concentrate on rethinking the model range.
Bohn’s efforts mainly produced an intense and protracted conflict over just what a Porsche should be. Widely divergent concepts were proposed, ranging from the revival of an “affordable” Porsche like the 914 and the 924 to a four-door model to be sold as an ultra-high-performance luxury sedan, to an even more performance-oriented Ferrari-type two-seater, following on the success of the 959 model in 1987. 7 In any case, new products were five or more years away due to the sequential nature of Porsche’s development system.
Because sales of the mid-priced 944 had collapsed after 1987 but demand for the more expensive 911 and 928 continued to be fairly stable until 1992, Bohn concluded that the mid-priced market ought to be left to the Japanese, with new Porsche offerings concentrated on the highest-priced segments of the market. In other words, Porsche was to pursue a classic segment retreat strategy, and the decision was finally made in 1990 to develop totally new two-door and four-door models with the engine in the front and rear-wheel drive to replace, by 1996, the 911, 928, and 944 and move Porsche further upmarket in price.
T ABLE 9.1: P RODUCTION H ISTORY OF P ORSCHE C ARS (000s)*
In the meantime, it seemed essential to cut the costs of production by about 30 percent to address the currency realignment between the dollar and the mark, yet no one inside the company seemed up to the task. The solution was soon found in thirty-eight-year-old Wendelin Wiedeking, the chairman of Glyco, a German auto parts maker. Wiedeking already knew the company and its problems because he had been manager of the paint and body shop at Porsche ten years earlier, before leaving for Glyco, where he had had great success, rising quickly to chairman, and had demonstrated both extraordinary energy and the courage to undertake dramatic change.
The Change Agent
Wiedeking arrived at Porsche in October 1991 as the sales slide was steepening and earnings were slipping from a meager $10 million profit in 1990– 91 toward a loss of $40 million in 1991–92 on $1.5 billion in sales. It was also just at the time that the Japanese car companies were launching their attack on German luxury cars and our MIT study, The Machine That Changed the World, was revealing to the Germans how far behind they had fallen in fundamental productivity.
However, Porsche’s problem was not primarily Japanese clones because even the “sportiest” Japanese cars, like the Toyota Supra and the Nissan 300ZX, were still several notches away in the direction of touring cars from Porsche’s pure “drivers’ cars.” Porsche’s fundamental problem was cost—its cars were simply too expensive for 1990s buyers to afford. And it was suddenly obvious that the amounts of time, effort, inventories, tools, and space needed by the best Japanese firms like Toyota to make “almost a Porsche” were a tiny fraction of those used at Zuffenhausen to make a real Porsche. It followed that costs and throughput times could be cut dramatically at Porsche if the right means were applied.
Wiedeking therefore called his direct reports together, had everyone read Machine very carefully, and arranged for an initial study tour in Japan. He remembers that the first shock was that the Japanese car companies they visited were willing to show them everything. “No one in the Japanese auto industry considered us serious competition and so they were very open. This was a major affront to our self-image.”
Upon their return, the team was terribly discouraged. “We could see that we were far, far behind and we had some general sense about why, but we lacked the techniques to tackle our productivity and first-time quality problems and we had no priorities. When you are way behind on every competitive dimension, how do you begin and where?”
Just then, at the beginning of 1992, the world recession caught up with sales of Porsche’s upmarket cars. Production at Zuffenhausen, which had rebounded in 1990 and 1991, suddenly fell by 23 percent from 26,000 to 20,000 and losses for the company as a whole were suddenly soaring past $150 million on only $1.3 billion in total sales.
Despite the growing sense of crisis, Wiedeking continued a series of trips to Japan, totaling four by mid-1992. These included managers but also shop-floor workers and members of the Works Council (the Metalworkers Union). He was intensely aware of the insularity of thinking at Porsche (which we believe is no worse than in the typical German engineering firm) and the need to open the windows.
Previously, Porsche operations managers had rarely traveled abroad, and then it was typically to look at high-tech machinery but not at management practices. This was on th
e premise that advances in management methods in foreign companies could not be relevant in Germany. The rank-and-file workforce and the union leaders had never been abroad on study tours and clung to a belief that all that was wrong at Porsche was a downturn in the market and some bad product decisions.
The Plan of Attack
As these visits continued, Wiedeking decided that he must take bold steps to dramatically reorganize the company, and that he must get help directly from Japanese experts, a decision he knew would be highly unpopular within Porsche. He already had a consultant working on a reorganization plan and he had met Maasaki Imai 8 of the Kaizen Institute when visiting Japan. In May 1992, Wiedeking invited the Kaizen Institute to work for Porsche as part of a four-pronged offensive to overcome the crisis.
The first step in the campaign was to restructure operations from six layers of managers to four (as shown in Figure 9.1 ) and to create four cost centers and three support functions to make responsibilities much clearer (as shown in Figure 9.2 ). The number of managers was reduced by 38 percent—from 362 in July 1991 to 328 in July 1992 to 226 by August 1993. In the new system, the support functions concentrated on developing the supply base, devising quality systems, and planning improvements while day-to-day operating tasks were assigned to the cost centers .
F IGURE 9.1: D ELAYERING O PERATIONS AT P ORSCHE
At the same time, Wiedeking negotiated with the Porsche Works Council for a new team structure on the plant floor. Production departments of twenty-five to fifty employees reporting through several layers of meister were broken down into two to three teams of eight to ten workers with each group of teams reporting directly to a single meister. (The ober meister and gruppen meister jobs were eliminated, as shown in Figure 9.1 .)
Wiedeking’s second step was a “quality offensive” to show the workforce the true costs of Porsche’s quality practices and to devise alternative methods. The most effective tool was to estimate the cost of catching a defect at the moment it occurred, compared with the cost at the end of the line, in the vehicle rectification area at the end of the plant, and in the hands of the customer. A problem costing 1 mark to fix at the spot it happened on the assembly line was estimated to cost 10 marks to fix at the end of the line, 100 marks in the vehicle rectification area at the end of the plant, and 1,000 marks at the dealer under warranty! This came as a revelation to the Porsche workforce, which had simply never looked downstream from their own work area to see the consequences of their mistakes.
A defect detection and reporting system was instituted so that everyone in every area of production could see immediately where mistakes were occurring and what was being done about them.
Wiedeking’s third step was in the form of a new suggestion system in which work team members were rewarded for submitting suggestions for improving both quality and productivity. The meister evaluated the suggestions immediately and took responsibility for implementing them quickly. Previously, suggestions had been sent to a special staff department from which they either never emerged or were seen only at a point so much later that the worker making the suggestion had moved on to a new job. As a result, the average employee made 0.06 suggestions a year.
F IGURE 9.2: N EW P RODUCTION O RGANIZATION
Under the new system, the number of suggestions per employee per year has risen to twelve, which is among the highest in Europe for European-owned firms. By contrast, the Lean Enterprise Research Centre survey of European auto suppliers in 1993 found that the average number of suggestions per employee per year at German-owned auto parts firms was less than one and that British-owned auto parts firms reported only two suggestions per employee. At the same time, Japanese auto parts firms in Japan reported twenty-nine. 9
The final step in the Wiedeking offensive was a policy deployment and visual management system called the Porsche Improvement (Verbesserungs ) Process, or PVP for short. This set measurable targets, monthly and annual, for each cost center and for each work team along four dimensions:
• cost, measured by reductions in hours of fabrication and assembly effort, and reductions in the amount of rework, scrap, and breakdown time for machinery.
• quality, measured as the number of first-time-through defects per component or per vehicle and defects discovered in the final road test of each vehicle .
• logistics, measured by on-time delivery to dealers, on-time delivery of parts to the next manufacturing operation, and reductions in inventory levels.
• motivation, measured by suggestions per employee, housekeeping, absenteeism, accidents, and PVP workshops and training hours per team.
When this system was launched—with great fanfare to coincide with the production launch of the 911 Carrera model in mid-1993 10 —each work meister ’s group agreed to the monthly and yearly targets for these measures and took responsibility for meeting them by posting their results prominently in their work area so that anyone walking by could see whether the team was succeeding. This was in complete contrast to the previous system, in which performance measures were secrets to be tightly guarded by top management and all proposals for improvement came from staff departments.
As the training progressed and it came time for the cost centers and work groups to take decisive steps to achieve their goals, Wiedeking was once more discouraged. He needed to introduce a total change in the thought process and practices of his craft-oriented workforce, yet he and his direct reports had only a theoretical knowledge of what to do. They had never actually implemented a lean system and the situation in the company was so desperate that they could not afford any initial failures. Wiedeking decided that Porsche needed shock treatment in the form of hands-on improvement activities from the Shingijutsu group he had met during his study tour of Japan. After several personal visits by Wiedeking and lengthy negotiations to prove Porsche was serious, Yoshiki Iwata and Chihiro Nakao agreed to take on the task.
The Arrival of the Sensei
As always, Chihiro Nakao’s initial foray into Porsche was a theatrical tour de force. When he arrived for his first visit in the fall of 1992, he insisted that Wiedeking immediately accompany him to the assembly plant. After walking through the door and looking at the stacks of inventory, he asked in a loud voice: “Where’s the factory? This is the warehouse.” Upon being assured that he was indeed looking at the engine assembly shop, he declared that if this was a factory Porsche obviously could not be making any money. And upon being told that Porsche was, in fact, losing more money every day, Nakao announced that a drastic improvement activity must be conducted in engine assembly along with many other places and that these must start immediately, indeed that day .
This, of course, was not the normal practice at Porsche, where all changes were carefully planned months in advance and negotiated with the Works Council. Any change in job content and the movement of any machine needed to be negotiated in advance, making kaikaku and kaizen in the normal “just do it” manner illegal in Germany.
Nor was it the normal practice for a stranger—a Japanese, no less, who spoke no German and communicated through an interpreter—to speak this way to a Dr. Ing. head of production (Ph.D. engineer) in a loud voice in front of the workforce. Finally, it was not normal practice to announce that the participants in the initial improvement projects must include all of the senior managers as well as the primary workforce.
The initial reaction on the shop floor was shock followed by considerable resentment, and the Works Council only very reluctantly consented to the initial improvement exercise. Most Porsche workers still found it difficult or impossible to believe that the problem was inside Porsche rather than outside in the marketplace. In addition, it was hard to believe that Japanese production engineers who knew nothing about the sports car industry could actually be helpful.
When the Works Council agreed to the experiment with the Japanese consultants, it stipulated that Porsche workers would conduct their own parallel workshop to show that if change was really needed it could
perfectly well be achieved by long-term employees rather than outsiders.
The objective of the first kaikaku in the engine assembly area was very simple: Get rid of the mountains of inventory and the treasure hunting for parts which occupied a substantial fraction of each assembler’s daily effort. Then make the parts flow from receiving to engine assembly to the final assembly plant very rapidly with no stopping, no scrap, and no backflows to fix defects.
A start must be made somewhere, so the objective of the first weeklong improvement activity was to cut shelf height in half from 2.5 to 1.3 meters in order to cut the inventory of parts on hand in engine assembly from an average of twenty-eight days to seven and to make it possible for everyone to see everyone else in the shop. (The underlying idea, of course, was to “lower the water level” so the snags in the prompt resupply of parts would be brought to the surface and the next step could be taken toward eliminating inventory and speeding flow.)
As the team formed its plan, a crucial moment arrived. Nakao handed a circular saw to Wiedeking, now dressed in the blue Porsche jumper worn by all production workers, and told him to go down the aisle sawing off every rack of shelving at the 1.3-meter level. As Manfred Kessler, then the head of the Methods and Planning Department and now the head of the Supplier Development Group, remembers, “It was the defining moment. Historically, senior management never touched anything in the plant and no one ever took such drastic actions so directly and quickly.”
Lean Thinking Page 24