This article was downloaded by: [University of Strathclyde] On: 25 November 2011, At: 06:59 Publisher: Taylor & Francis Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK Production Planning & Control Publication details, including instructions for authors and subscription information: http://www.tandfonline.com/loi/tppc20 Implementing the Lean Sigma framework in an Indian SME: a case study M. Kumar a , J. Antony b , R. K. Singh c , M. K. Tiwari d & D. Perry a a Division of Management, Caledonian Business School, Glasgow Caledonian University, Cowcaddens, Glasgow G4 0JF, UK b Centre for Research in Six Sigma and Process Improvement (CRISSPI), Caledonian Business School, Glasgow Caledonian University, Cowcaddens, Glasgow G4 0BA, UK c Department of Production Engineering, Birla Institute of Technology, Mesra, Ranchi-835 215, India d Department of Forge Technology, National Institute of Foundry and Forge Technology, Ranchi-834 003, India Available online: 21 Feb 2007 To cite this article: M. Kumar, J. Antony, R. K. Singh, M. K. Tiwari & D. Perry (2006): Implementing the Lean Sigma framework in an Indian SME: a case study, Production Planning & Control, 17:4, 407-423 To link to this article: http://dx.doi.org/10.1080/09537280500483350 PLEASE SCROLL DOWN FOR ARTICLE Full terms and conditions of use: http://www.tandfonline.com/page/terms-and-conditions This article may be used for research, teaching, and private study purposes. Any substantial or systematic reproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in any form to anyone is expressly forbidden. The publisher does not give any warranty express or implied or make any representation that the contents will be complete or accurate or up to date. The accuracy of any instructions, formulae, and drug doses should be independently verified with primary sources. The publisher shall not be liable for any loss, actions, claims, proceedings, demand, or costs or damages whatsoever or howsoever caused arising directly or indirectly in connection with or arising out of the use of this material. ProductionPlanning&Control, Vol.17,No.4,June2006,407–423 Implementing the Lean Sigma framework in an Indian SME: a case study M. KUMARy, J. ANTONY*z, R. K. SINGHx, M. K. TIWARI{ and D. PERRYy yDivision of Management, Caledonian Business School, Glasgow Caledonian University, Cowcaddens, Glasgow G4 0JF, UK zCentre for Research in Six Sigma and Process Improvement (CRISSPI), Caledonian Business School, 1 Glasgow Caledonian University, Cowcaddens, Glasgow G4 0BA, UK 1 0 xDepartment of Production Engineering, Birla Institute of Technology, Mesra, Ranchi-835 215, India 2 er {Department of Forge Technology, National Institute of Foundry and Forge Technology, b m Ranchi-834 003, India e v o N 5 2 9 5 6: Lean and Six Sigma are two widely acknowledged business process improvement strategies 0 at available to organisations today for achieving dramatic results in cost, quality and time by ] focusingonprocessperformance.Lately,LeanandSixSigmapractitionersareintegratingthe e d twostrategiesintoamorepowerfulandeffectivehybrid,addressingmanyoftheweaknesses y cl and retaining most of the strengths of each strategy. Lean Sigma combines the variability h at reduction tools and techniques from Six Sigma with the waste and non-value added Str elimination tools and techniques from Lean Manufacturing, to generate savings to the f bottom-line of an organisation. This paper proposes a Lean Sigma framework to reduce o y the defect occurring in the final product (automobile accessories) manufactured by a sit die-casting process. The proposed framework integrates Lean tools (current state map, 5S er System, and TotalProductive Maintenance (TPM))within Six Sigma DMAIC methodology v ni toenhancethebottom-lineresultsandwincustomerloyalty.Implementationoftheproposed U framework shows dramatic improvement in the key metrics (defect per unit (DPU), process [ y capabilityindex,meanandstandarddeviationofcastingdensity,yield,andoverallequipment b d effectiveness (OEE))and asubstantial financial savings isgenerated bythe organisation. e d a o Keywords: Six Sigma; Lean;Case study;Framework; SME nl w o D 1. Introduction the market share and maximise profit. All the large companies such as Toyota, Danaher Corporation, The last two decades has witnessed an increased General Electric, Motorola, Honeywell, and many pressure from customers and competitors for greater others, have achieved dramatic results by implementing value from their purchase whether based on quality, either Lean or Six Sigma methodologies in their faster delivery, or lower cost (or combination of both) organisation (Womack and Jones 1996, Harry 1998, in both manufacturing and service sector (Basu 2001, Basu2001,Murmanetal.2002,Sharma2003,Arnheiter George 2002). This has encouraged many industries to and Maleyeff 2005). adopt either Six Sigma (as their process improvement ShahandWard(2003)accentuatedtheimportanceof and problem solving approach) or Lean Manufacturing plant size, plant age, and union status on the likelihood (for improving speed to respond to customer needs and ofimplementing22manufacturingpracticesthatarekey overallcost)aspartofmanagementstrategytoincrease facetsoftheLeanproductionsystem.Thecorethrustof Leanproductionisthatitworkssynergisticallytocreate *Corresponding author.Email: [email protected] a streamlined, high quality system that produces ProductionPlanningandControl ISSN0953–7287print/ISSN1366–5871online(cid:1)2006Taylor&Francis http://www.tandf.co.uk/journals DOI:10.1080/09537280500483350 408 M. Kumar et al. finished products at the pace of customer demand with theintegratedapproachofLeanandSixSigmabusiness little or no waste. Lean strategy brings a set of proven strategies is delineated in figure 1. Figure 1 is based on tools and techniques to reduce lead times, inventories, the previous works of experts in Lean and Six Sigma set up times, equipment downtime, scrap, rework and (Womack and Jones 1996, James-Moore and Gibbons other wastes of the hidden factory. 1997, Hoerl 1998, Rother 1998, Breyfogle III 1999, The statistically based problem solving methodology Harry and Schroeder 1999, Emiliani 2000, Hines and of Six Sigma delivers data to drive solutions, delivering Taylore 2000, Pyzdek 2000, Antony et al. 2003, Snee dramatic bottom-line results. Linderman et al. (2003) and Hoerl 2003). The use of the comprehensive set of accentuated the importance of social and psychological tools mentioned above can help to reduce all kinds of considerations in understanding Six Sigma phenomena waste (rework, over production, waiting, material, and how effective use of goals helps alter the behaviour human skills, transportation and unnecessary move- of organised members and their perceptions about how ment)fromtheorganisation(Ohno1988,Womacketal. muchchangeispossible.Authorsfurtheremphasisedon 1990, Shingo 1992, Hines et al. 1998, Liker 1998). the understanding of the technical aspect as well as This paper presents a case study undertaken by behavioural insight for successful deployment of Six implementing a Lean Sigma framework into an Indian 1 Sigma. Sodhi and Sodhi (2005) illustrated how a global small- to medium-sized enterprise (SME) in order to 1 20 manufacturerofindustrialequipmentappliedSixSigma reduce the defects which occur in the final product er to its price setting process for one product line. In this manufactured by the company and thus satisfy their b m study, Six Sigma has transformed the tenor of the customers. The company was regularly receiving com- e v relationship between the pricing and sales staffs from plaints from its customers on crack propagation in the o N adversitytorelativeharmonyformakingjointdecisions automobile accessories manufactured by the company. 5 2 thatarealignedwithstrategicobjectivesofthebusiness. This was the major cause of customer dissatisfaction 9 5 Each methodology proposes a set of attributes that and was putting customer loyalty at risk. To retain its 6: 0 are prerequisites for effective implementation of the customers, the management realised the importance of at respective program: top management commitment, removing operational inefficiencies and wastes from the ] de cultural change in organisations, good communication organisation.Thegoaloftheorganisationwastoreduce y cl down the hierarchy, new approaches to production and the defects in the product, work-in-process inventory, h at to servicing customers and a higher degree of training scrap and rework cost. r St and education of employees (Salzman et al. 2002, Therestofthepaperisarrangedasfollows:Section2 of Antony et al. 2003). providesaninsightintothedie-castingprocess,theprob- sity Companies across the spectrum have found the most lem encountered by the company and the rationale er effective way to eliminate the flaws that lead to rework behind selecting Lean Sigma methodology to tackle v ni and scrap, and create one unified idea of continuous the problem. The reasons for using Lean Sigma as a U improvement, is the integration of Lean Manufacturing continuous improvement methodology for this case [ by and Six Sigma (Smith 2003). The integration of the two study is cited in section 3. Section 4 elucidates the ed systems can achieve much better results than either stepsinvolvedinimplementingtheproposedframework d a systemcanachievealone.While,Leanstrategiesplayan to identify the root cause of the problem and propose o nl important role in eliminating waste and non-value- correctiveactionto minimise theimpactof theproblem w o added activities across the organisation, Six Sigma, on customer satisfaction. The effectiveness of proposed D throughtheuseofstatisticaltoolsandtechniques,takes Lean Sigma framework is discussed in section 5. an organisation to an improved level of process perfor- Section 6 throws light on some of the difficulties mance and capability. The two methodologies empha- encountered while implementing the proposed frame- sisetheunfathomableinvolvementoftopexecutivesand work. The paper is concluded in section 7. communication with the bottom line to develop robust products and processes in their organisation. Most companies using the integrated approach apply 2. Company background basicLeantoolsandtechniquesatthebeginningoftheir program,suchascurrentstatemap,basichousekeeping Thedie-castingunitunderstudywasestablishedin1978 using 5S practice, standardised work, etc. After imple- with 150 employees, which comes under the category of menting the above tools and techniques some wastes SME as per the classification given by Indian Trade are eliminated from the system. Now, the tools and Industry. The organisation is engaged in designing and techniques of Six Sigma are used to offer powerful manufacturing various types of precision machined solutionstochronicproblems.Thecomprehensivesetof components using pressure and gravity die-casting tools,techniquesandprinciplesthatcanbeemployedin processes. The main customers of the company are Implementing the Lean Sigma framework in an Indian SME 409 Kanban DMAIC methodology Workplace management Variability reduction Set-up reduction time Statistical process control Total productive maintenance 5 why Process capability analysis Cause and effect Mistake proofing 5S practice Belt system (MB, GB, BB, YB) Pareto analysis Measurement system analysis Visual management Change management tools Design of experiment Robust design Value stream mapping Histograms Quality function deployment Control charts 11 Take time analysis Scatter diagram Failure mode and effects analysis 0 r 2 Just-in-time Project management e b m Production flow balancing Regression analysis e ov Kaizen N Analysis of means and variance 5 Cellular manufacturing 2 9 Hypothesis testing 5 LEAN SIX SIGMA 6: 0 ] at Figure1. The toolsand techniques ofLean andSix Sigma. e d y cl h at r ordinancefactories,theautomobileindustry,andtextile from where it is sent to the customer according to an St f machine manufacturers. The company manufactures agreed schedule. Customer orders are taken care of o y around250000units ofdiecastingproductsperyearto on the basis of first come first serve (FCFS). Quick sit caterfortheneedsofitscustomers.Theemployeeswork turnaround orders (QTA) are taken care of by r e v inthreeshiftsperday,eachshiftof8hours,andsixdays rescheduling the batch processing as decided by the ni U a week to meet the market demand. production manager. [ y The die-casting process starts with placing Al-alloy b ed (AlSi9Cu13)ingotsinthefurnaceandheatingthemfora ad sufficientduration.Whenthemetalmeltsandachievesa 3. Rationale for implementing Lean Sigma o nl suitabletemperatureinthecastingfurnace,itisinserted w o intothediesbyplungerpressure.Asthemetalsolidifies The application of different quality programmes to D the cast product is taken out with the help of an ejector reducetheoperationalinefficienciesandwastesrequires pinandplacedinatrolley.Thecastproductthengoesto topmanagementinvolvementandcommitmentinorder the trimming and fettling shop where extra projections to provide appropriate resources and training. The top are removed. The trimmed product is moved to the levelmanagementintheSMEwereproponentsofLean drilling section (MC1) where the different holes and andusedtopracticeTPM,Kaisen,and5Ssystemsinthe grooves are made as per the dimensions in the drawing. organisation. Management showed confidence at the In the next step, semi-finished products go to the beginningoftheinitiativeandsupportedthequalityand de-burring unit (MC2) where the external and internal production managers with the variety of resources and holes are cleaned and burrs are removed. It is then training required for successful deployment of Lean moved to the chamfering and threading unit (MC3) principles. As a result, there was less work-in-process where fine cutting at different angles along the surface inventory with reduced scrap and rework cost. and the making of external and internal threads are Management interest in the quality initiatives under- performed. Cleaning and polishing operations are per- taken started waning as the demand from its customers formed subsequently in the next stage. Finally, the increased. The wish to maximise return on investment finished product is stored in the dispatch department and the fear of not meeting the customer demand 410 M. Kumar et al. compelled the management to concentrate more on overall complexity, and helps to uncover the value- productionthanonqualityoftheproduct.Thisresulted added activities of a process, Six Sigma can solve in an increase in work-in-process inventory, scrap and complex cross functional problems where the root rework cost, and more defects (external and internal causes of a problem (in this case, crack propagation) casting defects like foliation, cracks, cold shut, pinhole are unknown and help to reduce undesirable variations porosity, etc.) in the final product. There were many in processes. The integration of two approaches elimi- hidden wastes embedded in the manufacturing process nates the limitations of individual approach. The team that were ignored by the company because their membersdecidedtoadoptSixSigmaoverTQMbecause manufacturing capacity was higher than their produc- of the following reasons: tion requirements. Problems were tackled by increasing . TQM focuses on long-term results and expected the work-in-process inventory leading to higher inven- pay-off is not well defined. Six Sigma creates a torycarryingcost.Inthelastsixyears,demandfortheir sense of urgency by emphasising rapid project productbecamehighduetoglobalisationandtheboom completion within 6 months. intheautomobilesector.Inordertomeetthecustomers’ . Improvement results from TQM initiatives are demand, production of automobile accessories was small and do not bring rapid changes. The die- 1 giventoppriority,irrespectiveofthequalityofproduct. 1 casting company under investigation is an SME 20 The management was able to meet the customer er demands by putting the quality of product at risk. and cannot afford to visualise tangible benefits b after a period of 1 year. m This resulted in a number of customer complaints from e . TQM provides a vast set of tools and techniques v different parts of the country. o withnoclearframeworkforusingthemeffectively, N Thisintimidatingsituationledmanagementtoponder 5 whereas Six Sigma uses DMAIC methodology for 2 over redeploying the quality initiatives taken at the 9 problem solving which successfully integrates a set 5 beginning. As most of the customer complaints were 6: of tools and techniques in a disciplined fashion. 0 related to crack propagation in the final die-casting at product (resulting in improper functioning of the . TQM is motivated by quality idealism, whereas ] Six Sigma is driven by tangible benefits for all de automobile engine), management formed a team to y major stakeholders (customers, shareholders, and hcl identifytherootcauseofproblems.Moreover,therewas employees). at a constant increase in in-process inventory, machine r St downtime,idletimeatdifferentworkstations,andthere It was decided that Lean would be used to create a of was also concern about the health and safety issues of foundation that allows the tools of Six Sigma to yield sity theemployeesastheaveragenumberofaccidentsonthe greater benefits, faster. Similar work was carried out er shop floor were increasing each year. by Cua et al. (2001), who investigated the practice of v ni The management, after a thorough brainstorming TQM and Lean simultaneously and the results revealed U session with the senior managers of different depart- that that manufacturing performance is associated [ by ments,generatedalistof15possibleprojectsthatcould with the level of implementation of both socially- ed enhance customer satisfaction. Multi-voting was then andtechnically-orientedpracticesoftheaforementioned d a used among its cross-functional team to narrow down programs. After a number of discussion sessions a o nl theprojectslisttoasmallerlistofthetopprioritiesorto framework was developed between the team members w o a final selection. Multi-voting allows the item that is and the facilitators, the latter being the authors of this D favouredbyall,butnotthetopchoiceofany,toriseto article.Thedetailsoftheframeworkarediscussedinthe the top (Tague 2004). In order to ensure greater next section. decision-accuracy, the voting was done by weighting the relative importance of each choice. Further, the outcomes from the multi-voting was debated by team 4. Framework for Lean Sigma implementation: members to avoid errors from incorrect information or a case study understanding about each project. One of the questions raised during brainstorming was related to the selection A model or framework is proposed by authors to of a continuous improvement method from a range of implement Lean Sigma in the organisation and is existing quality improvement programmes. The team delineated in figure 2. The framework is developed by decided to implement the Lean Sigma methodology to authorsafteranumberofmeetingswithtopandmiddle eliminatedefects,reducevariationandreduceinventory level management. The facilitators meticulously studied and overall complexity from the system. While Lean the whole die-casting process, met up with the employ- streamlines processes and eliminates waste (idle time, ees working on the shop floor and enquired about machine downtime, in-process-inventory), reduces the key parameters associated with each process, Implementing the Lean Sigma framework in an Indian SME 411 Steps in define phase (cid:127) Management initiatives (cid:127) Problem definition (cid:127) Brainstorming (cid:127) Develop big picture map (cid:127) Project charter Steps in measure phase (cid:127) Define performance standards (cid:127) Measurement system analysis DEFINE (cid:127) Monitoring the process (cid:127) Establish process capability Steps in control phase (cid:127) Control chart plotting CONTROL (cid:127) Share the lessons learnt MEASURE (cid:127) Mistake proofing exercise (cid:127) Sustainability plan ANALYSE 1 1 IMPROVE 0 2 er Steps in improve phase Steps in analyse phase mb (cid:127) Design of experiment (cid:127) Pareto analysis e (cid:127) Screen potential causes (cid:127) Select CTQ characteristics ov (cid:127) Discover variable relationships (cid:127) Cause and effect diagram 25 N (cid:127)(cid:127) EEssttaabblliisshh ionpge 5raSt isnygs tteomlerances (cid:127)(cid:127) BIdreanintisftyo rvmariinagtion sources 59 (cid:127) Implementing TPM 6: 0 at Figure2. Proposedframework for Lean Sigmaimplementation in the organisation. ] e d y cl h at scrupulously studied the documents of the production spent many hours on the shop floor observing, in order r St and quality departments to check on the defective to collect data and understand the different processes of products manufactured each day. This helped develop associated with the die-casting unit. y sit the Lean Sigma framework for implementation on the er shop floor. In the proposed framework, Lean tools are 4.1.2. Problem definition. A number of brainstorming v ni used within the Six Sigma (DMAIC) problem-solving sessions of team members were conducted to identify U [ methodologytoreducethedefectsoccurringinthefinal critical to quality (CTQ) characteristics based on the by product. voice of customer (VOC) input. In the meeting the ed problem of the die-casting unit, the size of the problem, d a the impact of the problem, etc., were discussed among o nl 4.1. Define the team members and it was apparent that most of the w o customercomplaintsrelatedtocrackpropagationinthe D 4.1.1. Management initiatives. An emergency meeting automobile accessories manufactured by the company. was called by top level management with operators, The goal of the team members was to identify the root engineers and senior managers of different departments cause of problem and reduce the defects occur in the to discuss the restructuring required in the current product. practices for enhancing the market share and customer satisfaction. In order to instigate enthusiasm and 4.1.3. Current state map. In order to have an insight motivation among employees for bringing about the into the current state of the die-casting unit, a current required changes, senior management decided to com- state map is developed which gives a closer look at the municate not only the successes but also the problems process so that opportunities for improvement can be encountered while implementing Lean Sigma frame- identified.Themovementofmaterialsthroughdifferent work.Thishelpedinidentifyingthemistakescommitted processes/facilitiesduringmanufacturingisshowninthe in other projects and learning from such mistakes. current state map (figure 3). A cross-functional team was formed consisting of Asshowninthecurrentstatemap,themanufacturing the operators, engineers from production, quality, and process starts with die casting and extends all the way marketing department, and senior managers. This team through to shipping. Below each process, process cycle 412 M. Kumar et al. ar e y mer es/ 14 Custo 250 000 piec Tray of Weekly/monthly PATCH 7 days S I D 0.4 day e ul 011 Forecast ping sched POLISH-ING/cleaning CT = 13 sec C/O = 6 m U/T = 90% NO. OF SHIFTS = 2DEFECT =0.2% 13 sec 101sec. mber 2 Ship 0.5 day ¼time e d 59 25 Nov Chamferingandthreadingunit CT = 14 sec C/O = 6 min U/T = 90% NO. OFSHIFTS = 3DEFECT =1% 14 sec Valueadde at 06: n 1 day days, sity of Strathclyde] Productiocontrol Weekly schedule De-burringunit CT = 17 sec C/O = 6 min U/T = 85NO. OF SHIFTS = 3DEFECT =0.5% 0.5 day 17 sec ¼p.Leadtime5.76 r a by [Unive Drillingunit CT= 14 sec C/O = 4 min U/T= 85% NO. OFSHIFTS = 3DEFECT = 2% 14 sec ntstatem d re ade 1 day Cur o nl 3. Dow 3 monthforecast Monthlyorder FETLING/Trimmingpress CT = 11 sec C/O = 20m U/T = 85%NO. OFSHIFTS = 3DEFECT =1.3% 11 sec Figure 2.36 days DIECASTING CT = 32 secC/O = 20 U/T = 90%NO. OFSHIFTS = 3DEFECT = 5% 32 sec er ur h y manufact Ingot ne mont Monthl ORE Al O ST Implementing the Lean Sigma framework in an Indian SME 413 time (CT), machine uptime (U/T), the number of shifts, 4.3. Analyse phase the changeover time (C/O), and the percentage defect Theobjectiveoftheteammemberswastodeterminethe are listed. It should be noted that data are collected rootcausesofdefectsandidentifythesignificantprocess basedoninteractionswiththeworkersatdifferentwork parameterscausingthedefect.Outofsixcastingdefects, stations. During the development of these state maps air inclusion, shrink holes, and gas holes porosity are it was found that the defect rate was high and was internal defects whereas cold shut, foliations, and unacceptable. soldering are surface defects. The internal defects are formedduringthecastingprocessasthemetalsolidifies. Themicroholescreatedinsidethecastingareduetoair 4.2. Measure phase or gas entrapment and result in crack propagation due to differential pressure and force created inside The team was divided into small groups to monitor the casting. This crack propagation impedes the proper the defects occurring in each process involved in the functioning of the final product and thus is very manufacturing of die-casting product. The team had significant to overall performance of the machinery been collecting data of defective products for the last 11 2 years and had identified the critical processes where where die-casting parts are fitted. The Pareto chart 0 showninfigure4illustratesthepercentagecontribution 2 maximum defects were occurring, but no action was r of internal and external defects in the process. It can be be taken. In order to validate the historic data, the team m concludedfromtable1andfigure4thatinternaldefects membersdecidedtocollectthedataofdefectiveproduct e v are the result of poor casting density and amounts to o for the following 6 days of production from their N 67% of total defects in the process. Other defects occur 5 respective work stations. The collected data was ana- 2 in the de-burring, chamfering and threading operations 9 lysed and it was found to match with the historic data, 5 due to tooling and clamping problems. 06: showing that the maximum numbers of defects were All the defects mentioned above decrease the sound- at coming from the die-casting machine, de-burring oper- ness of the casting, i.e. decrease the density of the ] ation (MC2), and chamfering and threading operation yde (MC3). casting. After conducting several brainstorming ses- hcl The next step was to determine a performance sions, the team members concluded that the density of rat standard based on customer requirements. A data the casting is the most important critical quality St characteristic in the die-casting process as it was related of collection plan was established to focus on the project to many internal defects (air entrapment, gas holes y output and also to carry out the standard setting sit exercise for the same. A Gauge repeatability and porosity, shrink holes, etc.). The objective of the die- er casting process was to achieve ‘better casting density’ v reproducibility (R&R) study was conducted to identify ni while minimising the effect of uncontrollable param- U the sources of variation in the measurement system and [ eters. To have a clear picture of the process parameters y to determine whether it was accurate or not. A study b d was performed to check the accuracy of gauges used e d for the measurement of characteristics as well as the a nlo reproducibility of the worker in performing operations Pareto Chart for Internal and External Casting Defects w on the machine. The Gauge R&R study performed on Do the system showed a variation of 8.01%, which implied 100 100 thatthemeasurementsystemwasacceptable.Usinghigh 80 80 performance and accurate equipment, we were able to secure the correct measurements of the die-casting parameter values and react in time for the necessary unts 60 60 cent correctiveactionsfortheexperimentalprocedure.What Co Per 40 40 the customers want is a sound casting with measurable characteristics, such as the density of the casting. 20 20 Therefore the ultimate goal of the team was to increase casting density. 0 0 Thecompanywasoperatingatabaselinecapabilityof Type of Defects Internal Defect Cold ShutFoliationSoldering Other Defects Count 67 20 7 3 3 0.12withdefectsperunit(DPU)being0.18.Thedesired Percent 67.0 87.0 7.0 3.0 3.0 Cum% 67.0 20.0 94.0 97.0 100.0 specification limit of casting density was 2.73–2.78g/cc and the casting produced before the implementation of Figure 4. Pareto chart for the internal and external casting Lean Sigma had an average density of 2.45g/cc. defects. 414 M. Kumar et al. Table 1. Classificationof defects andtheir contribution to totaldefect. Totaldefects from 5/01/04–10/01/04 De-burringunit—195 Chamfering andthreading unit—81 Casting—177(allcasting defects) Castingdefects—160 Other defects—35 Castingdefects—65 Other defects—16 Defects dueto poor casting—402(outof 523) Internal defect External defect Airinclusion 50 Cold shut 10 Shrinkholes 80 Foliations 20 Gas holes 130 Soldering 12 Porosity 90 Other defect 10 Total 350 52 Percentage ofinternal defect—67% oftotal defect 1 1 0 2 r e b m e SHOT SLEEVE v o MACHINE N 5 2 9 Filling timing 5 6: Length 0 Fast shot set point ] at Plunger velocity (2nd) Diameter e d y cl Plunger velocity (1st) Filling level h at Pressure Lubricant r St Casting f o density y Gate rsit Lubricant Temperature e v ni Cooling system U y [ Venting system Composition b d e d a o nl DIE METAL w o D Figure 5. Cause andeffectdiagram. affecting the density of casting, a ‘cause and effect’ At this stage, it was essential to identify significant diagram was constructed and is shown in figure 5. parameters so that they are tuned properly to achieve The cause and effect diagram shows that the most the desired range of casting density. important process parameters that affect the casting density are: piston velocity at first stage, piston velocity at second stage, metal temperature, filling time and 4.4. Improve phase hydraulic pressure. From experience, it was revealed that non-linear behaviour of the parameters of the die- 4.4.1. Design ofexperiment. Intheimprovephase,the castingprocesscanonlybedeterminedifmorethantwo team decided to carry out a designed experiment to levelsareused.Theparametersalongwiththeirsettings identify the significant process parameters affecting the are given in table 2. casting density. The most appropriate orthogonal array Implementing the Lean Sigma framework in an Indian SME 415 Table 2. Process parameters withtheir rangesandvaluesat threelevels. Parameter destination Process parameters Range Level 1 Level 2 Level 3 A Metal temperature((cid:2)C) 610–730 610 670 730 B Pistonvelocity 1ststage (m/s) 0.02–0.34 0.02 0.18 0.34 C Pistonvelocity 2ndstage (m/s) 1.2–3.8 1.2 2.5 3.8 D Filling time(ms) 40–130 40 85 130 E Hydraulicpressure (bar) 120–280 120 200 280 Table3. Results of L OA. 27 Trial no. A B C D E A(cid:3)B A(cid:3)C B(cid:3)C R1 R2 R3 Average S/N 1 1 1 1 1 1 1 1 1 2.336 2.338 2.441 2.372 7.500 2 1 1 2 2 2 1 2 2 2.339 2.442 2.447 2.409 7.637 1 3 1 1 3 3 3 1 3 3 2.442 2.505 2.448 2.465 7.839 1 0 4 1 2 1 2 2 2 1 2 2.427 2.444 2.416 2.429 7.713 2 r 5 1 2 2 3 3 2 2 3 2.545 2.577 2.595 2.572 8.210 e b 6 1 2 3 1 1 2 3 1 2.435 2.336 2.374 2.382 7.538 m e 7 1 3 1 3 3 3 1 3 2.716 2.728 2.701 2.715 8.680 v o 8 1 3 2 1 1 3 2 1 2.346 2.429 2.392 2.389 7.566 N 9 1 3 3 2 2 3 3 2 2.439 2.442 2.445 2.442 7.759 5 2 10 2 1 1 2 3 2 2 1 2.445 2.501 2.487 2.478 7.884 9 11 2 1 2 3 1 2 3 2 2.439 2.441 2.398 2.426 7.701 5 6: 12 2 1 3 1 2 2 1 3 2.418 2.381 2.443 2.414 7.658 ] at 0 1134 22 22 12 31 12 33 23 23 22..544529 22..541633 22..540445 22..542506 87..083018 de 15 2 2 3 2 3 3 1 1 2.543 2.585 2.591 2.573 8.212 cly 16 2 3 1 1 2 1 2 3 2.441 2.493 2.502 2.479 7.887 h 17 2 3 2 2 3 1 3 1 2.594 2.588 2.591 2.591 8.274 rat 18 2 3 3 3 1 1 1 2 2.539 2.542 2.545 2.542 8.108 St 19 3 1 1 3 2 3 3 1 2.474 2.495 2.489 2.486 7.914 f o 20 3 1 2 1 3 3 1 2 2.603 2.595 2.588 2.595 8.288 y rsit 2212 33 12 31 21 13 31 23 32 22..473084 22..467835 22..465922 22..465944 78..860131 e v 23 3 2 2 2 1 1 1 3 2.640 2.682 2.654 2.659 8.497 Uni 24 3 2 3 3 2 1 2 1 2.703 2.698 2.691 2.697 8.623 y [ 25 3 3 1 2 1 2 3 3 2.671 2.679 2.685 2.678 8.562 b 26 3 3 2 3 2 2 1 1 2.726 2.717 2.720 2.721 8.699 ed 27 3 3 3 1 3 2 2 2 2.745 2.747 2.752 2.748 8.785 d a o nl w o D (OA) design to meet the experimental requirement trial conditions. The average values of the S/N ratios is a 27-trial experiment (L OA) and the exper- foreachparameteratdifferentlevelsforallthetrialsare 27 imental layout is depicted in table 3. The company listed in table 4 and plotted in figure 6. The influence was initially operating with the following settings: of interactions on the casting density was negligible A , B , C , D , E . based on the analysis and was thus omitted from 1 1 1 2 3 The casting density is a ‘larger the better’ type the table and figure. of quality characteristic. Thus, the S/N ratio used is From figure 6, it is clear that casting density is at given by: maximum when the process parameters A, B, D and E are kept at level 3 and parameter C at level 1. Once the " # 1Xn (cid:1)1(cid:2) optimum settings of process parameters were identified, S=Nratio¼(cid:1)10 log ð1Þ n y2 the team members decided to implement 5S system and i¼1 i totalproductivemaintenance(TPM)toestablishaclean wherey isthecastingdensityforatrialcondition.Each environment within the shop floor and also to reduce i trail condition was repeated three times (i.e. n¼3). the idle time of machine and employees on the shop The S/N ratios are computed for each of the 27 floor.