NUMMI â€“ new united motor manufacturing
NUMMI is an automobile plant in Freemont, California, a joint venture between General
Motors (GM) and Toyota. It is sited on a plant which GM had closed in 1982 because of poor
quality. Productivity was among the lowest of any GM plant in the United States: absenteeism
was running at around 20 per cent, labour relations were dreadful, with wildcat strikes, and
alcohol and drug abuse. Soon after GM had closed the plant, agreement was reached with
Toyota to reopen it producing a Japanese-designed car, sold under the GM name, but
manufactured using Toyotaâ€™s methods of production. Over the next two years the plant hired
more workers, about 85 per cent of whom had worked in the plant previously. However, the
performance of the NUMMI plant could hardly have been more different. The plantâ€™s
productivity was more than twice as high as when it was run by GM, almost as high as Toyotaâ€™s
Takoaka plant in Japan. Quality also improved dramatically. Audits showed that, quality levels
were almost as high as Takoakaâ€™s and certainly higher than any other GM plant. Absenteeism
had dropped from over 20 per cent in the old GM-run plant to between three and four per cent.
Among the reasons for the success of the NUMMI plant were clearer organizational goals,
a selective approach to recruiting, and single status for everyone in the factory, even the pride of
working on a better designed product. However, the new plant and its management did not
abandon the techniques of scientific management which the previous plantâ€™s regime had
supposedly used. The philosophy of job standardization was still rigorously applied. Every job
in the plant is carefully analyzed using method-study principles to achieve maximum efficiency
and quality. Jobs are timed, using stop watches, and the detail of jobs questioned critically. Yet
whereas before, the companyâ€™s industrial engineers were in charge of applying method-study
techniques, now it is the operators (or team members as they are called) themselves who
perform the analysis of their own jobs. Team members time each other, using stop watches, and
analyze the sequence of tasks in each job. They look for alternative ways of doing the job which
improve safety and efficiency and can be sustained at a reasonable pace throughout the day.
Each team will then take its improved job proposals and compare them with those developed by
the comparable team doing the same job on a different shift. The resulting new job specification
is then recorded and becomes the standard work definition for all staff performing that job.
There were also several further benefits:
Â• Safety and work-related stress injuries improve because potentially dangerous or harmful
elements have been removed from the job.
Â• Productivity improves because wasted elements of the job have been eliminated.
Â• Quality standards improve because potential â€˜fail pointsâ€™ in the job have been analyzed out.
Â• Flexibility improves and job rotation is easier because standards are clearer and all staff
understand the intrinsic structure of their jobs.
One team leader compared the way in which the industrial engineers in the old plant had
designed jobs with the way it was done under the NUMMI regime. â€˜I donâ€™t think the industrial
engineers were dumb. They were just ignorant. Anyone can watch someone else doing a job and
come up with improvement suggestions â€¦ and itâ€™s even easier to come up with the ideal
procedure if you donâ€™t even bother to watch the worker at work, but just do it from your office
â€¦ almost anything can look good that way. Even when we do our own analysis in our teams
some of the silliest ideas can slip through before we actually try them out â€¦ thereâ€™s a lot of
things that enter into a good job design â€¦ the person actually doing the job is the only one who
can see all factorsâ€™.
1. What do you see as the main differences between traditional work study as described in the
text and the way in which NUMMI operates it?
2. What other aspects of job design seem to be put into practice at NUMMI?