Creating a lean supplier network: a distribution industry case

1We are indebted to Jon Hughes of ADR for introducing us to thisterm.

*Corresponding author. Tel.: #44 (0) 1222 874544; fax:#44 (0)1222874556; e-mail: [email protected].

European Journal of Purchasing & Supply Management 4 (1998) 235—246

Creating a lean supplier network: a distribution industry case

Peter Hines*, Nick Rich, Ann Esain

Lean Enterprise Research Centre, Cardiff Business School, 45 Park Place, Cardiff, CF1 3BB, UK

Abstract

This paper describes the application of a lean logistics approach to the development of a supplier network around a prominentdistributor of electronic, electrical and mechanical components. This involved mapping the activities of the firm, identifyingopportunities for improvement and then undertaking an improvement programme with the firm. The resulting Supplier Associationprogramme involved around 50 key suppliers across eight product category areas. The paper explores the different methods employedtogether with the support structure that was created. It concludes with evidence of the early results of the programmes as well asa number of key learning points for other organisations wishing to follow a similar path. ( 1998 Elsevier Science Ltd. All rightsreserved.

Keywords: Lean; Supplier association; Value stream mapping

1. Introduction

The management of change within a supply chain isperhaps the key role of the enlightened logistics, purchas-ing and supply manager. However, the task is not easy asthe manager is faced with a plethora of fads and ‘‘cure-all’’ solutions such as TQM, BPR and Lean Thinking.The problem here is that ‘‘these fads can, if not properlyunderstood and rigorously analysed, be implementedincorrectly or out of context and can do far more harmthan good2the major reason why fads fail is that thosewho rush to implement new concepts and ideas often failto understand that the relationships which work success-fully in one business environment may not be as success-ful when transplanted elsewhere’’ (Cox, 1995, pp. 72, 73).Cox goes on to challenge the academic community to‘‘[develop] a coherent body of empirically verified know-ledge concerning which types of sourcing and supplyrelationships might best be used under what circumstan-ces, and with what costs and benefits to business profit-ability’’ (Cox, 1995, p. 73).

Inherent in this challenge are three themes. The first isthat different types of purchasing or procurement arerequired in different circumstances. This line of argument

may be termed contingent prescription1 as a wide range ofoptions (contingencies) are open in how to manage yourpurchasing (a prescription of a functional activity). Thesecond theme is that any one approach is unlikely to beapplicable broadly in a wide variety of circumstancessuch as between an automotive manufacturer and a ser-vice based organisation. Unfortunately, for the empiricistthis is difficult to disprove without a wide range ofevidence from a variety of industrial settings. The thirdtheme is that there will certainly be a purchasing orprocurement function.

However, it is the authors’ belief that the Cox logicmay be somewhat flawed as what may be termed a pre-scriptive contingency approach can, within its philosophi-cal parameters, find wide application. To explain further,such an approach would at its starting point define thegeneral philosophy or mind set to be applied such asa Lean Enterprise (Womack and Jones, 1994; Diman-cescu et al., 1997). It is therefore prescriptive at the frontend initiation. It would then adopt a very differentmethod of application (contingently) within this, concen-trating on key processes such as order fulfilment, supplierintegration and new product development (Dimancescuet al., 1997). The latter approach does not necessarily

0969-7012/98/$19.00 ( 1998 Elsevier Science Ltd. All rights reserved.PII: S 0 9 6 9 - 7 0 1 2 ( 9 8 ) 0 0 0 1 5 - X

involve the management of purchasing or logistics in thetraditional sense at all because it may not define theseareas as key processes to the company and its extendedenterprise or supply chain.

The purpose of this paper is to explore whether theprescriptive contingency approach to the application ofthe Lean Enterprise does indeed have some merit. Asnoted above, due to the nature of contingent prescriptionchallenge, it will never be possible to disprove Cox’scontention. However, if empirical evidence can be pre-sented from an environment and sector far removed fromthat in which the original source model was developedsome doubt can be placed on the contingent prescriptionapproach.

2. The lean enterprise

The term ‘‘lean’’ was popularised by Womack et al.(1990) as embodying a system that uses less of all inputsto create outputs similar to the mass production systembut offering an increased choice to the end customer. Thelogic behind lean thinking is that companies jointlyidentify the value stream for each product from conceptto consumption and optimise this value stream regard-less of traditional functional or corporate boundaries.This is done by teams organised between functions andbetween companies, supported by the relevant functionalspecialists. In order to facilitate this change process,it is necessary to define corporate strategy and to identifykey customer facing processes such as order fulfilmentand new product development together with key non-customer facing processes such as supplier integrationor environmental control. Once this is complete, rolesand responsibilities can be defined and appropriate struc-tures for improvement put in place (Dimancescu et al.,1997).

Although the lean approach was developed throughthe study of the Japanese automotive industry, it is claim-ed that its approaches and way of thinking can transcendcultural and industrial divides (Womack and Jones,1996). However, to do this the specific approach requiredwill differ as suggested by the prescriptive contingencyapproach described above. In order to test the efficacy ofthis approach it was decided to attempt to apply a LeanEnterprise approach in an environment far removedfrom the Japanese car industry.

The choice for empirical testing was the UK division ofa leading distributor of electronic, electrical and mechan-ical components and instruments. The firm has a UKrange of over 70,000 products which it sells from its twomain warehouses based in central England as well asa network of regional ‘‘over-the-counter’’ operations. Itdispatches more than 15,000 orders per day and employsmore than 3000 people in its UK operations. For thesake of clarity we will call the company Partsco.

3. Designing the change programme

At the outset of the change process it was decided by theresearch team and company alike that a Lean Enterpriseperspective would be adopted in the work. It was nothowever defined where in the company or extended enter-prise the work would take place. This was to be decided onthe basis of need. In order to identify where the needexisted, considerable time was spent with the companyinvolving a range of senior staff. This group representeda range of different functions in the business. The focushere was on unstructured or semi-structured interviewswith representatives from purchasing, inventory control,goods inwards, warehousing operations, marketing, qual-ity and finance. As a result of these discussions it waspossible to identify the strategy being enacted by Partscoin terms of market definition, expansion path, warehous-ing location and growth path, service requirements ofcustomer base and their internationalisation programme.

This information was vital in understanding not onlywhat the company was wishing to do but also what theirkey customer facing and non customer facing processeswere. It was possible to identify the following four keyprocesses within the business:1. Order fulfilment covering all the activities required from

the receipt of an actual order from a customer to itsultimate delivery the next morning, at the customer.

2. New product introduction covering all the activitiesrequired to bring a new product to market. Newproducts were generally offered to Partsco by a manu-facturer or distributor who already supplied themwith other products.

3. Sales acquisition involving all activities required tomarket the existing product range and actually obtainthe order from the customer.

4. Supplier integration involving all activities required tocoordinate the various suppliers’ activities withPartsco including the reduction of inter-company andintra-company waste (Hines, 1994a).A review of these key processes showed that the com-

pany was exceptionally good at the first and third withnew product introduction being highly effective althoughnot as rapid as Partsco would like and hence the com-pany had already identified this opportunity and set upan improvement programme.

The present case study therefore focuses on the sup-plier integration process which was identified as the keyprocess that was at the time underdeveloped and couldpotentially yield significant benefits if addressed.

4. Mapping the value stream

4.1. Scope

In order to identify the opportunities for improvementof the supplier integration process it was decided by the

236 P. Hines et al. / European Journal of Purchasing & Supply Management 4 (1998) 235—246

Table 1The seven wastes

1. Overproduction2. Waiting3. Transportation4. Inappropriate processing5. Unnecessary inventory6. Unnecessary motions7. Defects

authors and the Partsco management team to map theactivities between the buyer and suppliers. In doing so itwas necessary to decide where the handover lay betweenthe supplier integration process and the customer orderfulfilment process. A decision was taken to divide theprocesses at the pick face where goods were available fordespatch to Partsco’s customers. In choosing whichproducts to map, it was decided to look at one productcategory but to look at products with different demandand revenue profiles. The company already defined all ofits products according to a Pareto-split (A, B, C for bothvolume of sales and a separate A, B, C for value of sales).As a result there were nine category areas ranging fromAA (high volume, high total value) to CC (low volume,low total value).

4.2. Organisation

In order to simplify data capture one product waschosen from each of the Pareto categories from oneproduct range area. This was particularly effective as thecompany had already organised its order fulfilment pro-cess along value stream lines, that is with mixed teams ofbuyers, inventory controllers and marketers co-locatedand controlling all activities for one of the 17 productarea. These cross functional teams were supported bya number of small competency centres focusing on pur-chasing, inventory control and marketing, who providedtraining and access to knowledge and new developmentsthat could be used by team members. The product rangechosen was lighting products as the earlier discussionswith the senior managers had identified this as a ‘‘diffi-cult’’ range due to its seasonal variability and fragility.

4.3. Approach

The value stream mapping approach was adopted (forfull details of the approach see Hines and Rich, 1997a).The method involves the identification of value addingand wasteful activities based around Ohno’s seven wastesas shown in Table 1 (Monden, 1993). This was achievedthrough structured interviews with the senior manage-ment team described above representing a wide spectrumof functional responsibilities. This wide range of viewsallowed for necessary cross checking, triangulation and

outlier identification. Each participant was given a de-scription of the seven wastes (based on Bicheno, 1994)and was asked to apportion 35 points to the wastesaccording to how important they felt that the wastes wereto Partsco.

Within this exercise participants were asked to givebetween 0 and 10 points to each waste with a total of 35points. The result of this exercise proved that unnecess-ary inventory was by far the largest waste. Indeed thecompany was achieving stock turns of only around 3 peryear at the start of the work in late 1995. In addition, dueto the rapid increase in sales such a situation would leadto the requirement for a new warehouse in the UK every5 years.

Applying a weighted average of the wastes it waspossible to identify which of the available mapping toolswould be most appropriate using the waste/tool correla-tion chart (Fig. 1). Five tools were selected and the resultsof their application are illustrated in Figs. 2—6 with onlyone product displayed for simplicity in Figs. 2—4 and anaverage or overview in Figs. 5 and 6. Some key learningpoints from the mapping exercise will be briefly sum-marise below. It should be noted that only about 50% ofthe learning points are drawn directly from the mapswith the other based on the researchers’ observationsmade during and after the mapping project.

The process activity map (Fig. 2) showed that for thevast majority of time products are within the definedsupply chain no value is being added and indeed a newterm ‘‘minutes per million (MPM)’’ was used to describethe operating time to total time ratio due to the extreme-ly low ratio found. In addition many delay points wereidentified in goods inwards but a very low product failurerate was identified at the incoming quality check. It wasalso noted that the inherent high urgency of productsinside Partsco was rarely found in the external supplychain.

The supply chain response matrix (Fig. 3) identifiedthat the supplier lead times were on average very longwith other internal lead times considerably shorter. Italso showed that there was a very high level of availablestock in the main storage area with other stock holdingsrelatively low. There also appeared to be no direct pull ofgoods from market needs rather an inventory based pushsystem was in place. This issue was further exacerbatedby the less even demand profile of product shipped to the‘‘over-the-counter’’ operations.

The demand amplification map (Fig. 4) showed thatthe company was good at forecasting the demand ofproducts with the forecast and actual demand curves inclose parallel. However, there was a fundamental disloca-tion between actual demand and actual purchases re-quired. In addition, the on-time delivery performance ofsuppliers was poor with often lengthy delays occurring.

The quality filter map illustrated (Fig. 5) was based onan average of the surveyed products. As can be clearly

P. Hines et al. / European Journal of Purchasing & Supply Management 4 (1998) 235—246 237

Fig. 1. The seven value stream mapping tools.

seen the quality of incoming products, especially allow-ing for the very fragile nature of the lighting productsmeasured, is very high and there is relatively little inter-nal damage. However, the service defects, mostly as a re-sult of late delivery are very considerable. Some of this isa result of the product suppliers’ performance but morewas attributable to problems caused by the third partypacker being used.

The decision point analysis map (Fig. 6) was used toidentify where customer product pull reached in the totalsupply chain. The dislocation point where customer pullmet supply chain push was found to be at the pick face,the point indeed where the researchers had earlier de-fined as being the cut off between the supplier integrationand order fulfilment processes. This decision point defini-tion was used to challenge the storage culture that wasprevalent in Partsco and the industry in general. Thequestion ‘‘What do you call a distributor with no stock’’was asked with most answers being something like ‘‘introuble’’ rather than perhaps a lean thinking response of‘‘slick’’.

The above maps together with a much more detailedanalysis of the existing situation was fed back to thesenior management team at Partsco. Following this feed-back a discussion was facilitated about how the companyand its supplier integration process could move forward.In order to aid the discussion a ‘‘chocolate box’’ ofpossible options correlated with the seven wastes wasused (Fig. 7). The result of this discussion was that it was

decided by the management team to trial the use ofa supplier association as a structural mechanism to facil-itate a supplier integration process and to make radicalimprovements in Partsco’s supply chain.

5. Setting up the support structure

For the purposes of this work the usual manufacturingdefinition of a Supplier Association (Hines, 1994b) wasmodified to fit the purpose required by Partsco. Thedefinition used was that ‘‘a Supplier Association isa group of companies, linked together on a regular basisto share knowledge and experience in an open andcooperative manner’’. In order to implement the pilotscheme and in preparation for a later roll out programmea three-tier system of management was put in place(Dimancescu et al., 1997). This cascading system (Fig. 8)was employed with as many of the original group in-volved in the mapping as possible.

At the highest level was the steering board led by thehead of the Strategic Purchasing unit. This group consis-ted of the top managers from across the business. A keyrole of this group was to ensure that a critical target wasset for the programme. This target was simply and unam-biguously set at a doubling of stock turns within a three-year horizon. In addition this group scheduled a series ofbimonthly meetings to measure the progress of the work,to ensure that adequate resources were being employed


Fig. 2. Process activity map.

and to make sure that nothing interrupted or disruptedthe Supplier Integration process. The steering groupmade no attempt to tell the process development orproduct teams how to achieve the outline target that hadbeen set but provided the support to help them.

The process development team led by one of the staffwithin the Strategic Purchasing unit was formed frommiddle managers from various competency centres in thebusiness. The team included representatives from inven-tory control, purchasing and a team leader for two of theindividual product group teams. A key responsibility ofthis group was to develop a set of measures and perfor-

mance gaps to motivate and guide the pilot product teamand subsequent roll out groups. The resulting measurestogether with initial data from the first pilot team areshown in Table 2. Around these goals the process devel-opment team designed the broad guidelines of an optimalsupplier integration process and facilitated its subsequentdissemination to later product teams. It also ensured thatties between these product teams were created so thatknowledge was shared between them of what did and didnot work well.

The measures chosen were not designed to be a simpleaudit of supplier performance but were more designed to


Fig. 3. Supply chain response marix.

Fig. 4. Demand amplification map.

be a gauge of the degree of Supplier Integration or whatlater became termed a Balanced Benefit Benchmark. Thiswas because some of the measures such as forecast varia-bility were designed to more be a measure of Partsco’s

performance. Others, like delivery on time were more aimedat suppliers’ performance with a third group includingaverage stock turns at Partsco a reflection of how effec-tively the total supplier integration process was working.


Fig. 5. Quality filter map.Fig. 6. Deceision point analysis.

Fig. 7. The partsco chocolate box.

Fig. 8. The three-tier system of management.


Table 2Balanced benefit benchmarking for Partsco pilot team

Metric Targetperfomance

Currentperformance

Gap

Average lead time 5 weeks 8.23 weeks 3.23Average stock turns 6 3.6 2.4Forecast accuracy 45% 38.2% 6.8%Forecast variability ratio 1 : 2 1 : 3.1 —Delivery on time 95% 74% 21%Delivery correct quantity 95% 62% 33%Rejections 1000 ppm 5314 ppm 4314Customer returns 10000 ppm 60815 ppm 50815

Table 4Top-rated joint improvement methods and inputs required

Improvement methods Inputs required

1 Co-managed inventory (CMI) Mostly supplier initiative2 Due-date perormance Mostly supplier; better information

from Partsco needed3 Milk round Mostly supplier; Partsco to

coordinate4 Self-certification Shared work5 Stabilise schedules Partsco’s initiative6 EDI (electronic data

interchange)Shared work

Table 3Proposed improvement methods

Rank Partsco’s suggestedimprovement methods

Suppliers’ suggestedimprovement methods

1 Self-certification Co-managed inventory (CMI)2 Due-date performance Due-date performance3 Vendor rating Milk rounds (delivery)4 Stabilise schedules Ship to stock5 Milk rounds (delivery) Stabilise schedules6 Comanaged inventory Standard packs7 EDI (electronic data

interchange)Standard product

information8 Replenishment modules Replenishment modules9 Hotline to suppliers EDI

10 Safety stock at suppliers Self-certification

6. The pilot supplier association

The individual product teams were made accountablefor attaining these targets with the process developmentteam providing support when required. The first of theproduct teams dealing with industrial control equipmentwas created in early 1996, with members representing theinternal value stream: a buyer, an inventory controllerand a product (marketing) manager. They were sup-ported by other individuals outside the process develop-ment team from other key support areas such as quality,goods inwards, accounting and packaging design. Theteam was led in this case by the buyer who had thegreatest knowledge of the suppliers. The reason thisproduct team was selected of the six teams that volun-teered was due to the enthusiasm of the industrial controlequipment team as well as the support given by theirteam leader.

The role of this team was to implement the supplierintegration process with a group of their key suppliers.Of the 32 suppliers in this product area, a strategic groupof seven was chosen on the basis of expenditure as well aspotential and previous proactive stance towards Partsco.The seven represented around half of the external expen-diture by the product group. The companies chosen werea mixture of European manufacturers and distributors ofnon-European products.

The seven companies were first brought together ata local hotel to Partsco in February 1996. Prior to thisthe team had undertaken a careful analysis of what theywanted to get out of the Supplier Association and howthey might achieve it using the Value Stream AnalysisTool [VALSAT] (Hines et al., 1998), a modification of thequality function deployment method (Mizuno and Akao,1994). The method consisted of the Partsco product teambrainstorming their wants from the programme and pri-oritising them. With the help of the process developmentteam they deliberated different approaches to theachievement of these desired outcomes. They developeda correlation of the improvements required and possiblemethods to use, with a range of scores given depending

on the closeness of correlation. They determined the totalusefulness for each method by multiplying its weightingby the correlation score and summing the results bycolumns.

The result was a ranking of the top ten Supplier Integ-ration methods to achieve the needs of the customer. Thiswas presented to the Supplier Association members atthe first meeting together with a briefing on the goals ofthe Supplier Association and how it could be used in thesupplier integration programme. During this event, andin more detail afterwards, individual suppliers wereasked to undertake a similar VALSAT exercise on whatthey wanted to achieve from the programme. This al-lowed for a balanced view between buyer and the sup-pliers before deciding what actually to do. The results ofthe exercise are shown in Table 3.

Although Partsco and the suppliers clearly had differ-ent requirements, there was a very close alignment ofsuggested improvement methods. Indeed, seven werecommon to both top ten lists. However, only six of thesewere feasible in the short/medium-term horizon. Of thesesome would require joint input and others individualwork by Partsco or the suppliers (Table 4).

The awareness raising stage complete, it was thennecessary to educate and share knowledge about how to


Fig. 9. A framework model for roll out of Partsco’s supplier integration process.

effect these improvements and craft an action plan tomeet the targets that had been set. This was done ina series of workshops held at rotating sites in early 1996.Participants drew up detailed plans permitting bothPartsco and the key suppliers to implement the necessaryimprovements.

After only a few months and a continued monitoring ofthe balanced benefit benchmarking results some rapidimprovements began to emerge. The first area to feelchange was the on time delivery performance from sup-pliers which quickly rose from around 74% to over 90%.In addition, lead time also fell from over 8 week toaround 7 and stock turns rose by 25% for the productsfrom the participating companies. It soon became obvi-ous that the pilot was starting to produce the desiredresults. However, it appeared that the obstacles to furtherrapid development mainly lay more at Partsco ratherthan at the suppliers. Examples of these includedPartsco’s inability to order or accept deliveries more thanonce per month per product item due to IT constraints.In addition a strengthening of the senior managementsupport was required so that the steering board could bekept better informed and provide greater assistance andpresence within the process. It was also increasinglyobvious that if the pilot scheme was to be rolled out toother product category areas a more formal cost/benefitanalysis would be required to ensure that initial decisionstaken ‘‘on faith’’ would actually produce the desiredpayback to Partsco.

7. The roll-out programme

The roll-out programme involved starting a secondproduct category Supplier Association in the summer of1996 in the connector area. The group followed a verysimilar pattern to the first and indeed came up with analmost identical set of programmes based on theVALSAT analysis. As a result of this the process develop-

ment team decided to develop a fully costed outlineprogramme that could be followed by subsequentgroups. Clearly, it was important to learn from the firsttwo pilot groups and produce a standardised approachto Supplier Integration while still allowing new groupsscope to fine tune the approach according to differingrequirement of their individual product category areas.

7.1. Costing the benefits

Based on detailed analysis of the first two groups a fivestage framework model was developed (Fig. 9). Themodel consisted of discrete target steps towards a fullyintegrated final vision. It was felt that until at least stage4 it would not be necessary to make significant changesto the suppliers manufacturing processes as there wereeasier to achieve and highly significant cost savings to bemade broadly within the logistics area as will be ex-plained below.

The awareness stage involved setting up some basicsystems at Partsco and the suppliers such as the use ofreturnable modules or plastic tote bins between the buyerand individual suppliers. This would remove the need forcostly double handling of products on delivery togetherwith the removal of the cost and environmental waste ofthe existing cardboard packaging. This first stage wouldalso be accompanied by the introduction of EDI linksand the development of self-certified product that wouldavoid the need for quality and quantity inspection atPartsco’s goods inwards area. This first stage representedapproximately where the first pilot Supplier Associationgroup had reached by early 1997.

The second stage called the enabled stage involveda further streamlining of the goods receipt and goodsinwards process at Partsco involving the bar coding ofthe returnable modules so that they would only requireto be bar code swiped on delivery before immediate putaway. At the same time further lead time reductions areenvisaged as the Supplier Association team move to


Table 5A benefits projection

Stages 1 2 3 4 5Aware stage Enabled stage Coordinated stage Pulled stage Integrated stage

Service level 96.5% 97% 97.5% 98% 100%Lead time 34 w/days 20 w/days 15 w/days 10.3 w/days 3 w/daysInventory 77 w/days 46 w/days 34 w/days 20.6 w/days 5 w/daysQuality of product 2500 ppm 2000 ppm 1500 ppm 1000 ppm 100 ppmDelivery 50000 ppm 20000 ppm 10000 ppm 10000 ppm 1000 ppmInventory saving 1.41% 3.31% 4.05% 6.13% 7.56%Service level saving 2.11% 4.21% 6.32% 8.42% 16.83%Expediting saving 0.01% 0.02% 0.02% 0.02% 0.05%Customer returns savings 0.20% 0.24% 0.25% 0.27% 0.53%Obsolete stock savings 0.10% 0.20% 0.30% 0.46% 0.56%Total cost saving index 3.83% 7.98% 10.94% 15.30% 25.53%Equivalent cost saving (£M) 8.85 18.43 25.27 35.35 58.97

Fig. 10. Costing the benefits.

focus on time compression. These first two stages will seeprogress continued to be monitored by the balancedbenefit benchmarking process.

The third stage of development called the coordinatedstage involve the development of a milk-round collectionscheme with the key suppliers hand in hand with a dra-matically increased frequency of deliveries. The fre-quency of deliveries for key suppliers at this stage willmove from monthly (or less frequently) to weekly (ormore frequently). This coordinated stage is a prelude tothe fully pulled stage where supplier demand is linked toPartsco’s customer demand by the use of a kanban pullsystem employing the returnable modules as physicalkanbans in the system. At the same time as this supplierswill be encouraged to start addressing their manufactur-ing processes by the application of value stream mappingwhich will help them pin-point key areas for radicalimprovement internally.

The last integrated vision stage appears at this pointa long way off but involves a further dramatic improve-ment in lead time, reduction in inventory within analmost stockless environment with demand on suppliersclosely linked to customer demand. This vision is forPartsco still several years away. However, a justificationfor the actualisation of this vision has been made by thebenefit model shown in Table 5 and illustrated in Fig. 10.

As can be seen in these illustrations the supplier integ-ration process will yield dramatic benefits to Partsco andtheir suppliers both in terms of improved customer ser-vice for Partsco’s customers, better quality products andconsiderably reduced operating costs.

The savings that Partsco will enjoy will be largelybased on two key factors. The first is a significant in-crease in customer service which will result in a reductionin lost sales and better retention of existing customers. Asnoted above this is a crucial area of Partsco. The secondkey area is the reduction of inventory and the consequentcost savings that will result from this. These savings are

split between the cost of financing the stocks and thewarehousing costs required physically to maintain thesestocks. In particular, the cost to Partsco of buildinga new warehouse to stock inventory as the sales turnoverrises is in excess of £50 million. The programme describedabove will allow Partsco to avoid such an expense thatthey otherwise would have incurred within the next twoto three years.

7.2. The new supplier associations

In order to achieve the vision it has been necessary toincrease dramatically the number of product groups andsuppliers taking part in the Supplier Association pro-gramme. As a result, during January and February 1997a further six groups were started in separate productgroup areas. Thus within a year of the start of the Sup-plier Association programme over 50 of the top suppliershave become involved out of a total supply base ofaround 1500. However, this small number are respon-sible for around 25—30% of Partsco’s total purchases.Once these groups have had time to consolidate a furtherwave of Supplier Associations will be started both in the


UK and abroad so that around 50% of Partsco’s sup-pliers by value will be directly encompassed in theirsupplier integration process. This further roll out is plan-ned from the autumn of 1997.

8. Discussion and conclusions

In little more than a year Partsco has developed froma situation where it had only a very limited supplierintegration process to a strong system with a well-charted future development plan that is self-sustainingwithout significant outside facilitation support. This situ-ation has been brought about by the application of a leanenterprise approach.

Of particular importance in the process has been theuse of value stream mapping to raise awareness of theneeds for supplier integration together with the use ofa multi-group supplier association. This approach hasallowed Partsco to work with suppliers both more quick-ly and more effectively. It has also allowed developmentwork to be carried out with competing suppliers, allow-ing in the early stages benchmarking data to be fed backusing disguised names for the suppliers involved.

The work has been described by one senior Partscoexecutive as ‘‘not rocket science but very effective2.andwho wants rocket science anyway’’. Another productgroup team leader, although strongly supportive allalong, was concerned that they may risk upsetting oralienating the suppliers by asking too much of them.However, one important part of the work has been toremove this perceived fear of upsetting suppliers. Theviews of the suppliers towards the work have been verypositive and in many cases the initiatives and drive iscoming from suppliers rather than Partsco. The pro-gramme has greatly improved communications betweenPartsco and their suppliers and has raised Partsco’spresence within suppliers’ thinking as well as the import-ance of supplier integration with Partsco’s senior staff. Asan unexpected side benefit, the performance of suppliersnot in the supplier association programme also appearsto be improving. This may be explained in two ways.First a spread effect is occurring as members of theproduct teams start informally to implement similar im-provements, if at a slower rate with non member firms.Second, as some firms may have initially been disap-pointed to have not been included in the activity theymay be putting in extra effort into their relationship withPartsco in order to try and become a member at a laterdate.

It is informative now to relate back to the earlierdiscussion about whether the lean paradigm can effec-tively be applied to a vastly different cultural and indus-trial setting and whether such a prescriptive contingencyapproach really is valid. The empirical evidence present-ed in this paper would suggest that Partsco have gained

very significant advantage through the application oflean thinking and stand to gain even more greatly in thefuture.

It is important to note that a very similar (prescriptive)approach was adopted at the front end of this improve-ment activity at Partsco to that applied within otherLean Enterprise transformation programmes. This ap-proach involved the application of a Lean Thinkingapproach within a value stream environment involvingprogramme control and policy deployment through athree-tier system of management (Womack and Jones,1996; Dimancescu et al., 1997). In addition, other tech-niques or methods, all part of the lean toolkit, wereadopted. These include the use of value stream mapping,the value stream analysis tool (VALSAT), the supplierassociation and benchmarking.

However, although the general philosophy taken andthe toolkit applied were prescriptive in nature, it shouldbe noted that these tools were indeed selected froma wider range of tools, with others such as levelled pro-duction, set up reduction and cellular manufacturingrejected even for manufacturing suppliers (Monden,1993). In addition, the particular application of the se-lected tools varies markedly from a typical approachwithin an automotive setting showing the contingencyapproach adopted at the operational level within thisprogramme. Indeed, the particular multi-group SupplierAssociation adopted is unique among the 40 SupplierAssociations applied in Europe, most of which are withinthe automotive sector (Rich and Hines, 1997). The reasonwhy such a contingency approach was employed wasthat it was fit for the purpose that was defined in thisparticular instance.

This point is further illustrated by the use of theVALSAT. The tool itself, as described earlier, is a modifi-cation of the quality function deployment approach firstdeveloped in the Japanese shipyards for product develop-ment. When first refined into the VALSAT approach themethod was applied in a textile industry case in Africa(Hines et al., 1997). In the present case it was used in theUK distribution sector. However, instead of being ap-plied as is usual by an expert or experts from a singlecompany it was applied by a customer organisation andseparately by a group of suppliers, with the results thencompared. Again such a contingent application, as far asthe authors are aware, is unique but ideally fit for pur-pose.

The same general contingent approach was also ap-plied to the benchmarking process applied by Partsco.The use of a traditional supplier only measurement style,as used in the Lean Enterprise Benchmarking Project(Andersen, 1992) for instance, was rejected. Instead,a more appropriate balanced and tailored approach wasadopted which sought more to measure the total rela-tionship and performance of the buyer and suppliersalike (Lamming et al., 1996).


This contingency point was further brought homewhen one of the authors undertook an informal interviewwith one of the participants in the first Supplier Associ-ation group. The Japanese owned company that suppliedindustrial control parts also makes a wide range of otherproducts including lifts, ships, pens and cars. The seniormanagers present who were British noted that they wereintrigued by the use of a Supplier Association withina service industry setting as the method was commonlyused by their own organisation in Japan but only bymanufacturing assembly plants with their componentsuppliers. However, they went on to comment ‘‘it doesappear to be yielding significant results here too’’.

Returning to whether the Lean Enterprise-based pre-scriptive contingency approach will cross cultural andindustrial divides, the empirical evidence presented herewould suggest, at least in this case, that the answer is yes.However, from one case, total universal applicabilitycannot be proved. At least with Partsco, the reason forsuccess appears to be dependent on the breadth of under-standing of the Lean Enterprise’s potential, the enthusi-asm and skill of those involved in its application anddegree of lateral thinking involved in its contingent ap-plication.

References

Andersen Consulting, 1992. The Lean Enterprise Benchmarking Pro-ject. Andersen Consulting, London.

Bicheno, J., 1994. The Quality 50: A Guide to Gurus, Tools, Wastes,Techniques and Systems. PICSIE Books, Buckingham.

Cox, A., 1995. Pro-activity, value engineering and strategic procure-ment management. In: Kemp, R., Lamming, R. (Eds.), Proceedingsfor the 1st Worldwide Research Symposium on Purchasing andSupply Chain Management. IFPMM/CIPS/PMAC/NAPM, Tempe,Arizona.

Dimancescu, D., Hines, P., Rich, N., 1997. The Lean Enterprise: Design-ing and Managing Strategic Processes for Customer-Winning Per-formance. AMACOM, New York.

Hines, P., 1994a. Creating World Class Suppliers: Unlocking MutualCompetitive Advantage. Pitman Publishing, London.

Hines, P., 1994b. Internationalization and localization of the KyoryokuKai: the spread of best practice supplier development. InternationalJournal of Logistics Management 5 (1), 67—72.

Hines, P., Rich, N., 1997a. The seven value stream mapping tools.International Journal of Production and Operations Management17 (1), 46—64.

Hines, P., Rich, N., 1997b. Purchasing structures, roles, processes andstrategy: is it a case of the tail wagging the dog?. Proceedings of the6th International Annual IPSERA Conference. T 5/3 1—17. Univer-sity of Naples ‘‘Frederico II’’, Ischia (Naples), Italy, 24—26 March.

Hines, P., Rich, N., Hittmeyer, M., 1998. Competing against ignorance:advantage through knowledge. International Journal of PhysicalDistribution and Logistics Management 28 (1), 18—43.

Jones, D., Hines, P., Rich, N., 1997. Lean logistics. International Jour-nal of Physical Distribution and Logistics Management, Lean Logis-tics Special Edition 27 (4), 153—173.

Lamming, R., Cousins, P., Notman, D., 1996. Beyond vendor assess-ment. Relationship assessment programmes. European Journal ofPurchasing and Supply Management 2 (4), 173—182.

Mizuno, S., Akao, Y., 1994. QFD The Customer-Driven Approach toQuality Planning and Deployment. Asian Productivity Organisa-tion, Tokyo.

Monden, Y., 1993. Toyota Production System: An Integrated Ap-proach to Just-In-Time, 2nd ed. Industrial Engineering and Manage-ment Press, Norcross, Georgia.

Rich, N., Hines, P., 1997. Supply-chain management and time-basedcompetition: The role of the supplier association. International Jour-nal of Physical Distribution and Logistics Management, Lean Logis-tics Special Edition 27 (4), 210—225.

Rummler, G., Brache, A., 1990. Improving Performance: How to Man-age the White Space on the Organisation Chart. Jossey-Bass Inc, SanFrancisco.

Womack, J., Jones, D., 1994. From lean production to the lean enter-prise. Harvard Business Review. 93—103.

Womack, J., Jones, D., 1996. Lean Thinking: Banish Waste and CreateWealth in your Corporation. Simon and Schuster, New York.

Womack, J., Jones, D., Roos, D., 1990. The Machine that Changed theWorld. Rawson Associates, New York.


Creating a lean supplier network: a distribution industry case

Documents

Transcript of Creating a lean supplier network: a distribution industry case