Lean Production and Manufacturing Performance Improvement in Japan, the UK and US 1994-2001

Abstract

This paper reports the findings of a longitudinal study into manufacturing performance, lean production principles and buyer supplier relations in the Japanese, US and UK automotive industries. A total of 26 first tier component makers in the three countries were subject to detailed benchmarking exercises in 1994 and in 1999-2001. In each exercise data on labour productivity and quality performance were obtained, along with a series of quantitative measures indicating the extent to which each plant conformed to 'lean production' principles. The results show that the Japanese plants improved their labour productivity by around 20 per cent between 1994 and 2001, whilst productivity in the US plants remained flat over the same period. All plants improved their quality performance during the period, but the Japanese plants retained their lead with an average external defect rate of 81 parts per million (ppm), compared to 111 ppm for the US plants and 416 ppm in the case of the UK plants. Measures of leanness in the supply chain (inventory levels, delivery frequencies and so on) should be sensitive to any weakening of the inter-firm relationships that have historically characterized the Japanese auto industry. These measures showed no evidence of such weakening, although qualitative evidence suggested that a polarization of the Japanese auto industry may be occurring under the influence of foreign capital, with independent firms such as Toyota and Honda (and their suppliers) retaining a stronger 'Japanese' character than their counterparts who have entered into equity relationships with non-Japanese companies.

Full text

LEAN PRODUCTION AND MANUFACTURING PERFORMANCE
IMPROVEMENT IN JAPAN, THE UK AND US 1994-2001
ESRC Centre for Business Research, University of Cambridge
Working Paper No. 232
by
Nick Oliver
Judge Institute of Management
University of Cambridge
Trumpington Street, Cambridge CB2 1AG
Email: n.oliver@jims.cam.ac.uk
Rick Delbridge
Cardiff Business School
Colum Drive, Cardiff, CF1 3EU
Email: delbridger@Cardiff.ac.uk
Harry Barton
Cardiff Business School
Colum Drive, Cardiff, CF1 3EU
June 2002
This Working Paper forms part of the CBR Research Programme on Industrial
Organisation, Competitive Strategy and Business Performance.

Abstract
This paper reports the findings of a longitudinal study into manufacturing
performance, lean production principles and buyer supplier relations in the
Japanese, US and UK automotive industries. A total of 26 first tier component
makers in the three countries were subject to detailed benchmarking exercises in
1994 and in 1999-2001. In each exercise data on labour productivity and quality
performance were obtained, along with a series of quantitative measures
indicating the extent to which each plant conformed to ‘lean production’
principles.
The results show that the Japanese plants improved their labour productivity by
around 20 per cent between 1994 and 2001, whilst productivity in the US plants
remained flat over the same period. All plants improved their quality
performance during the period, but the Japanese plants retained their lead with
an average external defect rate of 81 parts per million (ppm), compared to 111
ppm for the US plants and 416 ppm in the case of the UK plants.
Measures of leanness in the supply chain (inventory levels, delivery frequencies
and so on) should be sensitive to any weakening of the inter-firm relationships
that have historically characterized the Japanese auto industry. These measures
showed no evidence of such weakening, although qualitative evidence
suggested that a polarization of the Japanese auto industry may be occurring
under the influence of foreign capital, with independent firms such as Toyota
and Honda (and their suppliers) retaining a stronger ‘Japanese’ character than
their counterparts who have entered into equity relationships with non-Japanese
companies.
JEL Codes: L2, L6, M1
Keywords: Lean Production, Suppliers, Auto Industry, Japan
Acknowledgements
The 1994 study was supported by Andersen Consulting. The 1999-2001 follow-
up study was supported by the UKs EPSRC. The authors are very grateful to
Professors Ikeda, Nakagawa and Ueda for their invaluable assistance with the
data collection in Japan in 2000-2001.
Further information about the ESRC Centre for Business Research can be found
on the World Wide Web at the following address: www.cbr.cam.ac.uk

Lean production and manufacturing performance improvement
in Japan, the UK and US 1994-2001
Introduction
During the 1980s and through much of the 1990s the performance
superiority of Japan’s automotive companies relative to their Western
counterparts was demonstrated repeatedly with respect to
manufacturing (Schonberger 1982; Womack et al 1990; Oliver et al
1994, 1996) and new product development (Clark and Fujimoto 1991;
Fujimoto 2000). “The “Machine that Changed the World” by
Womack, Jones and Roos was published in 1990 and has sold
hundreds of thousands of copies. This book ascribed Japanese
manufacturing superiority to “lean production” principles, a distinct
approach to the management of manufacturing centred largely on the
Toyota Production System (TPS) the principles of which, it has been
argued, are applicable to many other business processes as well
(Womack and Jones 1996).
However, prolonged recession in Japan has taken its toll and Japan is
no longer regarded as the economic paragon it was assumed to be 10
to 15 years ago. The reasons for this are not difficult to see. A number
of Japanese financial institutions have collapsed, amidst widespread
publicity. Several Japanese auto companies have experienced
financial problems and have entered into tie-ups with foreign auto
companies. The most public example of such a tie up was the merger
between the French car company Renault and Nissan, but many other
examples abound – Ford has had a substantial stake in Mazda for
many years, but has increased this recently; GM has a moderate stake
in Fuji Heavy Industries (Suburu); and Mitsubishi Motors are now
part of the Daimler-Chrysler Group. Only Honda and Toyota continue
to go it alone, though both have had, and in Toyota’s case continue to
have, joint venture activities with non-Japanese partners. Whereas in
the 1980s and early 1990s much attention in writings on the auto
industry focused on operational issues (manufacturing and new
product development) more recently the focus of debate has centred
more on strategic issues – acquisition and merger, systems supply,
1

modularity, e-procurement and exchanges and so on. Also, with the
efficacy of the whole Japanese system called into question (Porter et
al 2000), interest in Japanese firms as an example of a superior
business model, to be emulated by the rest of the world, has declined.
Consequently, performance comparisons between Japanese firms and
their Western counterparts attract less attention than they once did.
This is in many ways unfortunate. Although the focus of attention on
Japan in the 1980s and 1990s may have been unduly operational, to
the neglect of strategic issues, operational issues are still important
and Japan still has lessons for the rest of the world. Furthermore,
Japan’s prowess for continuous improvement has been one of the
features that delivered the manufacturing performance superiority so
graphically illustrated by the studies of the 1980s and 1990s; has this
been sustained in Japan, and to what effect? Japan’s inter firm
networks, once seen as a crucial source of Japan’s competitive
advantage, have been put under strain by prolonged recession and the
influx of foreign capital. How are these faring in today’s
environment? Commentators such as Williams et al (1994) have
argued that the success of Japan’s auto industry, and Japan’s
particular social arrangements of production, were a function of an
unusual set of historical circumstances. When long term growth came
to an end, Williams et al predicted, Japan’s car companies would start
to behave in ways indistinguishable from their Western counterparts.
Is this occurring?
The purpose of this paper is to address such issues via an analysis of
relative levels of performance – and rates of improvement – amongst
first tier automotive component manufacturers in Japan, the UK and
the US. The study on which the paper is based is unusual in that
detailed benchmarks of practice and performance were taken from a
panel of plants in the three countries in 1994 and again in the period
1999-2001 when the measurement exercise was repeated with the
same plants. This provides a rare glimpse of change over time at an
unusually high level of detail.
2

Companies and Methods
As Table 1 shows, data were collected from 35 plants in the three
countries in 1994 and from 29 plants in 1999-2001, of which 26
plants were common to both studies. The 1994 study also covered an
additional 36 plants in France, Germany, Italy, Mexico and Spain, but
these have been excluded from the analysis as we have no data on
these plants for 1999-2001.
The data collection process involved an initial visit to each plant by
members of the research team. At this visit the research team
introduced the project, carried out an inspection of the plant and
briefed the plant management on how to complete the questionnaire.
The questionnaire was then left with the plant for a period of four to
six weeks. This was a substantial document, with approximately
1,000 data fields, and typically took several days of management time
to complete. The research team then made a second visit to each
plant and reviewed the completed questionnaire with the plant
management, a process which could take anything up to one day.
Data collection during 1999-2001 followed exactly the same process.
In most cases, staff turnover in the plants meant that the research team
were dealing with a different set of respondents to those involved in
the 1994 study.
The questionnaire covered seven main areas: plant performance; plant
characteristics; process control; work organization; problem solving
and improvement; relations with suppliers; and relations with
customers. The main purpose of the questionnaire was to yield data
that would permit systematic comparisons of performance between
the plants in each product area, and profile the management practices
of each plant to ascertain the extent to which lean production
principles were in use. The questionnaire was constructed around the
model of lean production presented in Figure 1.
Plant performance was measured by physical productivity, in terms of
units per labour hour. This was calculated by taking the annual units
of output of each facility and dividing it by the annual hours of labour
3

input. Adjustments were made for vertical integration, for the length
of the working day, for overtime, absenteeism, and product
complexity (for exhaust plants only). Quality was measured as
defective units in parts per million (ppm) as reported to the plants by
their customers (the car makers) over the same twelve-month period.
The measures of management practice represented quantitative
indicators of the use of lean production principles. The `leanness' of
factory operations was measured by counting the hours of inventory
of specific parts at various stages along the production process.
Continuous improvement activities were measured by asking about
the presence or absence of suggestion schemes, the number of
suggestions per employee and the use, membership, and activity rates
of problem-solving groups such as quality circles. A similar approach
was applied along the supply chain, on both the customer and supplier
sides. Thus, data were gathered on the inventories of raw materials
and finished goods, on delivery frequencies both by suppliers and to
customers, on information exchange, joint problem solving activities
between firms and so on.
Results
The results are presented under the following headings:
1. Manufacturing performance
2. Context – volumes, headcount, product variety
3. Problem solving and improvement
4. Buyer-supplier relations.
Manufacturing Performance
The first question that we set out to address was the relative
productivity of plants in the three countries. Womack, et al (1990)
claimed a 2:1 gap between Japanese vehicle assembly plants and their
Western counterparts. Previous studies in the auto components sector
(Oliver et al 1994, 1996) showed smaller gaps than this, although
these were still substantial and virtually always in favour of the
Japanese plants.
4

Table 2 demonstrates that the Japanese plants continue to outperform
both the US and British plants in terms of labour productivity
(minutes per unit). The Japanese seat plants are twice as productive as
their US counterparts. The exhaust and brake plants show seven per
cent and 15 per cent performance differentials respectively in favour
of the Japanese over US plants. Overall, the British plants show an
even greater productivity shortfall vis a vis the Japanese.
When the 2001 figures are compared to the 1994 performance figures
for the same plants, a surprising picture emerges. The Japanese plants
average an increase in labour productivity of 20 per cent, whereas
labour productivity in the US plants has remained more or less static,
and the British plants actually show a decline of 13 per cent. The net
impact of this is a widening of the productivity gap between the
Japanese plants and the US and British plants between 1994 and
2001.
The pattern of performance differentials is repeated with respect to
defect rates. The Japanese plants average an external defect rate of 81
parts per million (ppm), some 25-30 per cent lower than the US plants
and one fifth of the rate of the British plants. Compared to 1994 levels
of quality, plants in all three countries show big improvements, in
particular the British plants, though in the British case this was from a
very high 1994 baseline of approximately 1,700 ppm. In 1994, US
and Japanese levels of defects were similar; six years later there are
signs that the Japanese may be opening up the gap again.
Context
What explains the apparent declines in labour productivity on the part
of the UK and US plants? Declines in production volumes that have
not been matched with a corresponding adjustment in staffing levels
are an obvious explanation, but on average production volumes have
risen, indicating that these plants are winners in terms of pressures for
consolidation of production in the auto industry. In absolute terms
production volumes in the Japanese and US plants were broadly
comparable, and roughly double those of the UK plants. However, it
5

is striking that the expansion of output of the Japanese plants between
1994 and 2001 (+33%) has been achieved with only a modest increase
to headcount, whereas the UK and US plants not only show
substantial increases in volumes but also show substantial increases in
numbers of employees. It should be noted that differences in vertical
integration, product complexity, overtime, absence and non-working
time all mediate in the relationship between headcount, volumes and
labour productivity, and that the changes in labour productivity are
not a straight function of changes to headcount and production
volumes.
Product proliferation, as indicated by increased numbers of live part
numbers, could be one explanation of static or depressed labour
productivity, with its implications of shorter production runs, more set
ups and associated logistics headaches. Fujimoto (2000) has reported
efforts to increase the use of common parts across different vehicle
platforms in Japan, and argues that this may be increasing the length
of time spent in the planning stages of the product development cycle.
The measure of product variety in this study was the number of live
part numbers. Changes on this measure may be driven by strategies
on the part of the component makers such as diversification of their
customer bases, as well as increases in product variety on the part of
the car makers that they already serve.
The Japanese plants show the greatest product variety, by a
substantial margin – more than double the level of the UK and six
times that of the US plants. As already demonstrated, production
volumes in Japan and the US are broadly comparable, suggesting
radically different volume/variety mixes in the two countries.
Moreover, when the 1994 and 2001 figures are compared it is clear
that the US plants are on a very different trajectory to the Japanese
and UK plants, managing to reduce product variety by over 50 per
cent whilst at the same time increasing volumes by about the same
amount. This suggests progress with parts standardization and
consolidation of production that is as yet absent in Japan or the UK.
6

Capacity utilization was also explored as a driver of changes in labour
productivity. The US showed a slight drop in capacity utilization, and
Japan and the UK showed increases in capacity utilization. This
suggests that some of the boost to Japanese productivity is due to a
better matching of output to capacity between 1994 and 2001.
However, capacity utilization provides no clue as to why productivity
of the UK plants has fallen.
Plants in all three countries show substantial falls in external defect
rates between 1994 and 2001. However, there are differences in the
patterns of reasons behind these defect rates, as Table 4 illustrates.
Amongst the Japanese plants, human errors in manufacturing stand
out as the single most common cause of defects, in contrast to the UK
and US where technical issues (for example machine problems) are
the most frequent explanation of defects. Suppliers also stand out as a
particular source of quality problems in the US, consistent with the
figures on defect rates of parts coming in from second tier suppliers,
shown in Table 6. In the 1994 study the US second tier performed
poorly relative to the first tier and this pattern does not seem to have
been addressed in the interim.
During the plant inspections in Japan, the most visible manifestations
of quality improvement efforts were techniques and devices to reduce
human errors – poke yoke. For example, since 1994 several plants had
introduced devices such as infra-red sensors across the openings of
line side bins that held small components such as fasteners, washers
and other fittings. These sensors detected whether an operator had put
his or her hand in the bin to pick up a fitting, and unless this had
occurred prevented the work piece moving on to the next workstation.
This reduced the probability of components being missed out of the
assembly process. One seat plant had taken this a stage further by
fitting covers to component bins – these covers were opened
automatically with the arrival of the work piece. However, a bar code
on the work piece controlled which covers were opened, thereby
7

eliminating the possibility of incorrect components being fitted to the
work piece.
Given such innovations, it is somewhat surprising to see ‘human
error’ appearing as the most significant cause of defects in the
Japanese plants; this may be a testament to the progress that has
already occurred in other areas, such as machine and supplier
reliability.
Problem Solving and Improvement
The literature has made much of the bottom-up problem solving
found in Japanese factories. This is manifested through employee
suggestion schemes, in problem solving groups such as quality
circles, and most generally under the generic banner of ‘kaizen’
activities.
Consistent with the patterns found in the 1994 study, the Japanese
plants continue to show most activity on our measures of kaizen
activity, as shown in Table 5. Japanese plants show the highest
participation of production operators in problem-solving groups and
the highest number of suggestions per head. Both US and UK plants
show substantial increases in suggestions per employee, but this is
from a relatively low base in 1994 and so the differential between the
US and UK plants and the Japanese plants remains substantial. There
was little change in suggestions per head in Japan over the period,
suggesting a ‘topping out’ at around 25 suggestions per employee per
year. In interviews, we explored the question of whether incremental
process improvement could continue to yield the saving necessary to
meet the cost reduction targets imposed by the car makers. Most
respondents felt that incremental process improvements of themselves
could not continue over long periods to deliver the required cost
reductions and looked to value analysis and value engineering
(VA/VE) techniques, and design-led cost reductions to drive out cost.
One respondent commented that incremental process improvement
over prolonged periods was akin to “trying to squeeze water from a
dry towel”.
8

Buyer-Supplier Relations
The relatively close relations between buyers and suppliers in the
Japanese auto-industry, manifested by the keiretsu system, have been
one of the most noted features of the Japanese system, and it has been
argued, provide a major support to both the manufacturing and
product development operations of the Japanese car companies by
creating long term, high trust relations that facilitate co-operative
behaviour such as joint cost reduction activities and problem solving
and permit the tight coordination for JIT principles to work along the
supply chain (Helper and Sako 1995; Lamming 1993; Macduffie and
Helper 1997; Nishiguchi 1994; Nishiguchi and Beadet 1998; Sako
1992). The prolonged recession in Japan may be expected to affect
some of these characteristics. A contraction in the market is likely to
place long term collaborative relationships between buyers and
suppliers under stress, as the game moves from being win-win to win-
lose. There are certainly some signs that this is occurring. For
example, in late 1999 Nissan announced that it was reducing its
number of suppliers from 1,145 to less than 600 by 2002, selling its
shares in all but four of its affiliate companies, and adopting Western-
style competitive bidding for new contracts amongst its suppliers.
Nissan parts suppliers reacted angrily to this:
“Nissan officials are shirking their responsibility for
having not been able to make cars that sell. Instead they
are blaming us suppliers”.
“We shall be forced to stop purchasing automobiles from a
company that coldly cuts us off” (Daily Yomiuri, 28
January 2000).
Ironically, the mentality of interdependence between companies,
banks and suppliers, which has been seen as a strength of the Japanese
business system, is now regarded as part of the problem, at least by
Nissan’s French partners. This had led some observers to argue that
the traditional Japanese business system is breaking down and
9

converging towards a more “Western” (usually conceived of as
market-based) model of buyer-supplier relations.
If this interpretation is correct, then the quantitative indicators of the
closeness of buyer-supplier relations in Japan should exhibit signs of
a loosening of relations. These figures are shown in Tables 6 and 7
(for links between the plants in the study and their suppliers) and
Tables 8 and 9 (for links to car makers).
In 1994, the Japanese plants had approximately double the number of
suppliers of their US and UK counterparts. Seven years later the
number of Japanese suppliers had increased by 11 per cent (mainly
due to increases in product variety). US plants showed much larger
increases in the number of second tier suppliers (50 per cent plus)
which is surprising given the US rationalization of product ranges
noted previously. One possible explanation is that the continuing
poor performance of the second tier in the US has forced the first tier
extend their supplier bases in the search for more competent suppliers.
This does not appear to have proved successful given the quality
performance figures in Table 6.
On the two main indicators of supplier performance, on-time delivery
and defect rates of incoming parts, Japanese plants continue to
outperform UK and US plants. Since 1994 the latter show some
improvement in terms of on-time delivery but virtually no change in
terms of defect rates, accounting for the relatively high incidence of
reports of 2
nd
tier supplier-induced defects in the quality performance
of the 1
st
tier plants.
The measures of inventories and delivery frequencies, which we use
to show the closeness of relations between the focal plants in the
study and their suppliers (at least in a logistical sense) continue to
show much tighter links between 1
st
and 2
nd
suppliers in Japan than in
the US or UK. There is a 1:5 differential between Japanese and US
plants and a 1:9 differential between Japan and the UK in terms of
inventory levels. Japan and the US both show comparable falls in
10

inventory levels over the seven year period, in the order of 25 per
cent.
Thus, there is little evidence of a loosening of relationships at the 2
nd
tier/1
st
tier interface in Japan during the period covered by the study.
It may of course be that the logistics-based indicators of closeness are
not sensitive to the changing commercial arrangements (such as the
awarding of contracts on the basis of price-based competition) that are
unfolding around them. Alternatively, it may be that any such changes
are more marked at the car maker/1
st
tier supplier interface where the
impact of influences such as foreign capital are most evident. The
results pertinent to this are shown in Tables 8 and 9.
The data on the car maker/1
st
tier interface show a similar pattern to
that already observed at the 1
st
tier/2
nd
tier interface. Japanese plants
show much better performance in terms of on time delivery (by a
factor of over 40), have far lower inventories and more frequent
deliveries compared to the UK and US plants. All these measures
indicate a much tighter coupling between car-makers and suppliers in
Japan than is found in the other two countries. Moreover, it can be
seen from Table 9 that the Japanese plants show more improvement
on these measures between 1994 and 2001. Of the three countries, the
Japanese plants show the least change in terms of numbers of car
makers that they serve, again indicative of stability and continuity,
rather than change and revolution, in Japanese buyer-supplier
networks.
Conclusions and Implications
What do these patterns of change in Japan, the UK and US
demonstrate? First, they show that, at least as far as these auto
component plants are concerned, Japan has not lost its edge over the
US and UK in terms of manufacturing performance. The Japanese
plants have continued to make improvements in terms of labour
efficiency and still lead their US and especially their UK counterparts,
by a significant margin. Labour productivity in the US and UK plants
has been more or less static (and actually shows a decline in the case
11

of the UK). Although on average production volumes in the plants in
all three countries have risen, Japan has managed this with a much
smaller increase in headcount than either the US or UK plants, and
without a major rationalization of product ranges. The US plants have
made considerable progress in rationalizing their product ranges.
Plants in all three countries have made significant progress in
reducing the proportion of defective products that reach their
customers (that is, the car makers, in the case of this study). Japan
continues to lead the US in quality performance by a margin of
around 30 per cent, whilst the UK trails a distant third. Defects in
incoming parts are a particular problem in the US, suggesting that the
manufacturing reform that has been occurring amongst the car makers
and first tier suppliers has still to penetrate the second and third tiers.
The continuing poor performance of US second tier suppliers is
striking. In 1994 we noted that US first tier suppliers were struggling
to cope with poor quality and delivery performance from their second
tier suppliers, and that their role as a quality ‘filter’ added
considerable strain and cost. These findings were acknowledged and
confirmed by the industry at the time but the subsequent period has
seen little improvement. One impediment to improvement in
suppliers’ performance appears to be that purchasing decisions are
taken primarily on a cost basis, by headquarters functions unfamiliar
with the operational and logistics requirements of their own plants.
Given the sustained, and in some areas increasing, performance
advantage of the Japanese plants, it is precisely in areas such as these
that renewed interest should be taken.
The measures relevant to the closeness of buyer supplier relations
largely present a picture of continuity, rather than change, in Japan.
The tight logistics symptomatic of close social relations between
buyers and suppliers have if anything become tighter over the last
seven years. Of course it may be that changes in the commercial
relations between firms do not affect such operational details, though
this would run counter to what has been the accepted wisdom through
12

much of the 1980s and 1990s, namely that it is the very existence of
tight social relations that permits and facilitates operational
excellence. An alternative explanation is that changes in social
relations are occurring, but that the lag inherent in any such changes is
obscuring this.
The results of this study also point to an apparent paradox. The
Japanese plants clearly perform very well operationally, but a more
macro economic analysis does not present the same picture of
success. The largest component firms in the world are predominantly
US and European firms – only one or Japanese firms figure in the top
ten. Similarly, many Japanese firms have been experiencing financial
problems of one sort or another and recourse to foreign capital has
been one response to this. Detailed treatment of this issue is beyond
the remit of this paper, but one interpretation of this is that Japanese
suppliers are suffering due to the structural features of the Japanese
auto industry – specifically a relatively large number of car and
associated component makers, with relatively high dependency
relations between car makers and their main component suppliers.
This structure facilitates cooperation between car makers and
suppliers, but means that the risks faced by the component makers are
relatively concentrated. In the event of a prolonged recession, as
Japan has faced, such a structure may more rapidly lead to financial
problems than one in which component makers can spread their risks
across a wider base of car makers, and possibly car-producing
regions. In this respect, the very conditions that encourage operational
excellence through greater intimacy between buyers and suppliers
may work against the spreading of risk – and vice versa.
This said, the economic problems experienced by Japan at a
macroeconomic level should not distract from the continuing lessons
that may be gleaned from operational assessments of Japanese
manufacturers. In difficult circumstances, Japanese plants have
continued to improve their operational performance. The concept of
continuous improvement is one of the most significant components of
the Japanese model of manufacturing; the evidence reported here
13

suggests that this concept remains an enduring feature of Japanese
manufacturers and it remains an area where Western manufacturers
may have much to learn.
14

TABLES AND FIGURES
15

TABLE 1: JAPANESE, UK AND US PLANTS (1994 and 1999-
2001)
1994 1999-2001 Number
Common to
Both Studies
Japan 9 10 8
UK 12 9 9
US 14 10 9
Total 35 29 26
16

FIGURE 1: THE LEAN PRODUCTION MODEL
Inside the Factory Along the Supply
Chain
Flow JIT, low inventories of
WIP, ‘pull’ systems of
production control,
simple work flow,
team based work
organization, visual
control
JIT deliveries, low
inventories of
incoming parts and
finished goods
Error Prevention High process control,
work standardization,
poke yoke, design for
manufacture
Joint planning, design
and development, high
visibility of processes
along the supply chain,
schedule stability, staff
exchanges
Improvement Problem solving and
continuous
improvement groups,
suggestion schemes
Joint problem solving
and cost reduction,
supplier associations
17

TABLE 2: PRODUCTIVITY, QUALITY AND CHANGE OVER
TIME 1994-2001
Japan UK US
Minutes of direct labour to
produce a unit, 2001:
Seat plants
Exhaust plants
Brake plants
45.5
7.5
5.5
83.3
9.8
13.8
90.1
8.1
6.5
Average change in labour
productivity 1994-2001
+20% -13% -2%
External defect rates (ppm,
2001)
81 416 111
Change in defect rate 1994-
2001
-58% -75% -35%
TABLE 3: CONTEXT AND CHANGE OVER TIME
Japan UK US
Headcount (direct and indirect) 275 240 306
Product variety (excluding
exhausts)
1
357 145 63
Change in production volumes
since 1994
+33% +62% +53%
Change in headcount since
1994
+11% +49% +55%
Change in product variety since
1994
+12% +45% -52%
1
Due to special circumstances (high aftermarket requirements,
options of shipping part-products this measure is prone to unreliability
for exhaust plants and has therefore been excluded)
18

TABLE 4: SOURCES OF DEFECTS
Japan UK US
Design issues 7.5% 10.3% 7.4%
Suppliers 8.0% 15.7% 25.0%
Manufacturing -
technical issues
7.9% 36.9% 41.9%
Manufacturing – human
issues
68.8% 33.5% 21.9%
Other 7.8% 3.6% 3.8%
TABLE 5: PROBLEM SOLVING AND IMPROVEMENT
Japan UK US
% of operators
involved in problem
solving groups
88.1% 70.0% 52.0%
Suggestions per head 24.5 1.9 4.0
Annual target per
operator
19.3 2.0 13.3
% of suggestions
from production
operators
69.0% 87.4% 42.3%
19

TABLE 6: SUPPLIER RELATIONS, 2001
Japan UK US
Number of suppliers 78 32 56
Incoming defect rate (ppm) 463 3,861 7,752
% of late deliveries from
suppliers
3.5% 4.4% 12.3%
Hours of incoming parts 10.6 93.8 55.4
Frequency of delivery from
suppliers (every x hours)
6.1 41.0 25.5
TABLE 7: INDICATORS OF CHANGE OVER TIME IN 1
ST
TIER/2ND TIER RELATIONS
Japan UK US
Number of suppliers +11% +7% +51%
Hours of incoming
parts
-26% -10% -25%
Frequency of
deliveries from
suppliers
+8% -4% +53%
20

TABLE 8: CUSTOMER RELATIONS, 2001
Japan UK US
Number of customers 3.0 2.4 2.3
% of late deliveries to
customers
0.1% 4.8% 4.2%
Finished goods inventory
(hours)
2.4 69.6 30.0
Frequency of delivery to
customers (every x hours)
4.1 15.5 10.2
TABLE 9: INDICATORS OF CHANGE OVER TIME IN 1
ST
TIER/CAR MAKER RELATIONS
Japan UK US
Number of
customers
+3% +42% +13%
Hours of finished
goods
-79% -1% +18%
Frequency of
delivery
+20% -30% +12%
21

References
Clark, K. and Fujimoto, T. 1991. Product Development Performance:
Strategy, Organisation and Management in the World Auto
Industry, Boston, MA, Harvard Business School Press.
Delbridge, R., Lowe, J. and Oliver, N. 1995. Measuring
manufacturing performance: The process of benchmarking.
International Journal of Production and Operations Research,
Vol 15, No 4, 50-62.
Fujimoto, T. 2000. Shortening lead time through early problem
solving – A new round of capability building in the auto industry.
In Jurgens, U. (ed) New Product Development and Production
Networks. Berlin, Springer.
Helper, S.R. and Sako, M. 1995. Supplier Relations in Japan and the
United States: Are They Converging?, Sloan Management
Review, Spring, pp. 77-84.
MacDuffie, J.P. and Helper, S. 1997. Creating Lean Suppliers:
Diffusing Lean Production Through the Supply Chain.
California Management Review, Vol 39, No 4, 118-151.
Nishiguchi, T. 1994. Strategic Industrial Sourcing: The Japanese
Advantage. New York and Oxford: Oxford University Press.
Nishiguchi, T. and Beaudet, A. 1998. The Toyota Group and the
Aisin Fire, Sloan Management Review, Vol. 40, No. 1, pp. 49-59.
Oliver, N. , Delbridge, R, and Lowe, J. 1996. Lean production
practices: international comparisons in the auto components
industry. British Journal of Management, Vol 7, No 1, S29-S45.
Oliver, N., Delbridge, R., Jones, D. and Lowe, J. 1994. World class
manufacturing: further evidence in the lean production debate.
British Journal of Management. Vol 5, 53-63.
Porter, M.E., Takeuchi, H. and Sakakibara, M. 2000. Can Japan
Compete? Palgrave, New York.
Sako, M. 1992. Prices, Quality and Trust. Cambridge: Cambridge
University Press.
22

23
Schonberger, R. 1982. Japanese Manufacturing Techniques. New
York: Free Press.
Williams, K., Haslam, C. Johal, S. and Williams, J. 1994. Cars:
Analysis, History, Cases. Providence: Berghahn Books.
Womack, J.P. and Jones D.T. 1996. Lean Thinking. New York: Simon
and Schuster.
Womack, J.P., Jones D.T. and Roos, D. 1990. The Machine that
Changed the World: The Triumph of Lean Production. New
York: Rawson Macmillan.