A COMPARATIVE STUDY OF ENGINEERING …

1 A COMPARATIVE STUDY OF ENGINEERING CHANGEMANAGEMENT IN three SWEDISH ENGINEERINGCOMPANIES1A COMPARATIVE STUDY OF ENGINEERING CHANGEMANAGEMENT IN three SWEDISH ENGINEERINGCOMPANIESP eter Pikosz and Johan MalmqvistMachine and Vehicle DesignChalmers University of TechnologyS - 412 96 G teborg, Swedentel: +46 31 772 1367fax: +46 31 772 1375e-mail: paper analyses the ENGINEERING change (EC)process in three different ENGINEERING companies inSweden. In the paper, the influence of various companyspecific factors, such as change leadtime, for the design ofthe EC process are analysed. The current use of computersupport is surveyed and the possibility to apply a modernproduct data management (PDM) system to support theprocess is analysed.

2 The paper also presents differentstrategies for improving the ENGINEERING changemanagement process as well as of the product datamanagement systems in order to achieve an :product data management , PDM, ENGINEERING change management , workflowmanagement, process management1 INTRODUCTIONThe hard competition on the marketplace demandsthat the companies can develop and manufacture complexproducts with a higher performance and quality to a lowerprice than before to stay competitive. The product life-cycles have decreased during the last period, whichemphasizes the ability to develop the products in a shortertime in order to maximize the market window. The oldsequential way of performing product development (PD)and produce large series of expensive products is moreand more turning into production of customer variantswith a fast adoption to changing customer needs,requiring that the EC process can be performed increasing product complexity utilizing multipletechnologies along with the shorter time for the PDprocess makes it necessary to have more people withdifferent competences involved in each PD project,forming a concurrent ENGINEERING trend for the industry is quite clear with moredigital product and production verification using virtualreality and digital mock-up technology.

3 This willinevitably lead to an increased amount of digital files,which in turn demands a secure product datamanagement, that can keep track on all virtual andphysical products and their corresponding documen-tation. This impacts on the data management within theEC process, since an increasing amount of documentshave to be managed and ideal when designing a product is to get thedesign right the first time, eliminating the needs to makechanges. However, changes due to wrong decisions andchanges in customer requirements are unavoidable. If aundesired design decision has been made the product datahas to be changed in a controlled fashion. The impact ofthe change has to be investigated and all documentsrelated to the error have to be identified and changed.

4 Inthe best case only some documents have to be changed,whereas the worst scenario is that the production has to bestopped and already delivered products have to berecalled with a substantial prestige loss for the changing approved documents, many companiesuse a formalized ENGINEERING change management (ECM)process where the important issues when making achange are pointed out. The process should be designed2to solve the problem and prevent new errors. An ECprocess that is fast and reliable can even improve thecustomer EC process requires extensive communicationbetween different functions within a on how far the product is in the life-cycle,more and more functions, ranging from design to after-sales, have been involved.

5 In order to provide the bestsolution possible, all involved functions must be able togive their opinion on the change . This is oftenaccomplished by a cross-functional ENGINEERING changecommittee, which decides upon the changes. The ECprocess does however not end with the committeedecision. The information has to be processed severaltimes through the company before the change can be fullyintroduced. Due to the intensive data management withinthe EC process, vendors of product data management (PDM) systems often provide an EC module (CIMdata,1995). These modules are based on the assumption thatthe EC process can be generalized and implemented in acomputer system. One of the purposes of this paper istherefore to determine whether the process can begeneralized or not based on the conditions in differentcompanies.

6 By modelling the EC process in threeengineering companies, specific company factors for thedesign of the EC process have been determined. Based onthe models, suggestions of improvements of bothprocesses and the PDM systems have been found in orderto be able to provide a good interaction between the ECprocess and the PDM outline of the paper is as follows: section 2 givesan introduction to ENGINEERING change 3 describes the method used in the case studies atthe companies presented in section 4, whereas the casestudies are in section 5, section 6 and section 7. Thestudies are compared and analysed in section 8, andfinally, the conclusions are presented in section change ECM?In many companies the formal use of the engineeringchange process starts when a design project has reachedits first baseline ( System Design Review ), afterwhich all project documentation will be submitted forapproval and locked for further work.

7 The baseline is acheck-point before starting new activities that maydemand accurate information ( detail design orproduction). In some industries ( the defence industry)the customer is involved in the baseline approving thework done up to this point. All the documents are verifiedagainst specifications and released. Once released,documents may only be changed by applying a formal reason for making changes can originate fromdifferent causes. The draft standard ISO11442-6 (1996)gives the following examples: change of a part depending on altered function orproduction requirements change in the application of a part introduction of a new part replacement of a part withdrawal of a part correction of errors on a document bringing an old document up-to-dateThe examples above are very much focused on partsand documents.

8 A view that is more focused on differentstages in the product development (PD) process woulddescribe the causes as: changes in the customer specification faults in the interpretation of customer demands intotechnical requirements difficulties in parts fabrication or assembly weaknesses in the product identified during prototypetesting quality problems with some subsystem or component development for future product revisionsIf the complete formal process would be used everytime a released document changes it would have a verylarge impact on the progress of the PD process. In orderto reduce the amount of changes companies often treatchanges that do not affect the form, fit or function of apart with a simplified EC process.

9 Another way to reducethe lead times for making changes is to classify changesin a more sophisticated way and treat them with differentprocesses depending on the nature and the importance ofthe change (Balcerak and Dale, 1992).The EC process is a core process of the largerconfiguration management (CM) process. Each change ofthe product or of its corresponding documents causes achange of the product configuration. This means thatevery time a new revision of a product document iscreated, it must be linked to the correct productindividuals produced. When a malfunctioning product isreported to the company, the corresponding documentrevisions must be easily found. This is particularimportant in the defence industry and when many productvariants are produced (Merlenbach, 1992).

10 Theimplementation of an efficient as well as effective ECprocess is thus of general importance for a sucessful with the EC processThe EC process is associated with certain formal use of the process starts when documents havebeen released, which means that the documents have beendistributed from ENGINEERING to other departments withinthe company. The reason to change the documents isoften discovered at other departments than different departments are affected when documentsare to be changed. In order to perform the change in thebest way possible and fulfil the wishes of all parties, theEC process should be performed in a cross-functionalenvironment (Diprima, 1982). This has often proven to bedifficult, since functions typically have different goal of ENGINEERING is to perform a certain function inthe best way possible, whereas the goal for manufacturingmight be to assemble in short time and cutting materialcosts might be the main objective for purchasing and EC process is often associated with long leadtimes, which is caused by a number of factors.