Government Documents

A Model for Product Risk Estimation Through Corporate Memory and Techniques Integration

A Model for Product Risk Estimation Through Corporate Memory and Techniques Integration
of 5
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Related Documents
  A model for product risk estimation through corporate memory andtechniques integration Vrassidas LEOPOULOS 1 Konstantinos KIRYTOPOULOS 1 Vaggelis BELLOS 11 Mechanical Engineering DepartmentNational Technical University of AthensIroon Polytechniou 9, 157 80 ZografouGREECE  Abstract: Aim of this paper is to enhance product risk estimation by providing an integrated model. Thesimilarity of “technical” and “managerial” risk management methods is identified and there is an effort tocompromise two of the most known and widely accepted ones, the FMECA (Failure Modes, Effects andCriticality Analysis) and the PMI (Project Management Institute) risk processes. Moreover, the paper suggeststhat one of the most important factors for an efficient risk management process implementation is themaintaining of a corporate memory. The information existing in the corporate memory feeds a risk estimationalgorithm, which provides a common, for “technical” and “managerial” risk management methods, index of the product risk. Key-Words: product risk, risk management, corporate memory, risk algorithm 1   Introduction This paper presents a model for product risk estimation. The concept that has been followed inorder the algorithm supporting the model to beproduced, was based on the similarities that havebeen identified in all basic and widely accepted risk management methodologies. Whether “technical” or“managerial” risk management methodologies arebased on the same principles thus they could bemerged. The methodologies that have beenexamined here are the FMECA (Failure Modes,Effects and Criticality Analysis) and the PMI(Project Management Institute) risk processes.The model, adopts the maintenance of a kind of memory, which is called corporate memory since itspurpose is to track the risks that have been identifiedin the past in a specific enterprise or field of business. The main goal is to arm the enterprise withthe knowledge and the experience acquired from thepast concerning the product risk.The rest of this paper is organised as follows: InSection 2 the problem of the study is formulated.The need of a common risk management process isargued and the basic methodologies that were usedare analysed. The method proposed for the solutionproblem is described in Section 3. The benefits fromthe use of a risk management corporate memory areenlighten and the proposed algorithm of the model ispresented. In Section 4 the conclusions andopportunities for further research in the topic arediscussed. 2 Problem Formulation Risk knowledge is fuzzy, unstructured, insufficient,tacit (in people minds only for example), forgottenas a solved ancient history, transformed afterwards,insufficiently organised and not or incompletelycatalogued, underestimated, registered inheterogeneous information systems in the company,or even secretive [1].However, risk is the key factor for product failureand should be confronted with great care andfurthermore, it should be regarded as a strategiccompetitive factor [3]. In this paper two of the most  representative methodologies for “technical” and“managerial” risk management, respectively, havebeen selected and are presented right afterwards. 2.1FMECA methodology As written by Dhillon [4], Failure Modes and EffectAnalysis is an outstanding tool for the design andanalysis of engineering systems. It may simply bedescribed as an approach to perform analysis of eachpotential failure mode in a system so as to study thepotential effects of such a failure to the system itself.The method is an enhancement of Failure Analysis,which has been developed by the U.S Navy in orderto establish a mechanism for reliability control overthe detail design effort [5]. Nowadays, FMECA iswidely used in aerospace, defense and nuclearpower generation, as well as in many other complexsystems, covering the technical part of risk management. Figure 1 presents the steps that shouldbe followed in order to perform the FMECA. 2.2PMI methodology On the other hand, each system or product containsa “managerial” risk apart from the pure technical Describe the system and itsassociated functional blocksIdentify failure modes andtheir associated effectsPrepare critical item listDocument the analysisEstablish ground rulesDefine the system and itsassociated requirements Fig. 1: FMECA process steps Risk ManagementIdentificationQuantificationResponseControl Fig. 2: PMI risk management steps one. Since the production of more and moreindustries becomes project based, probably the bestmethodology to be followed in order to cope withthis type of risk is the one that is proposed by theProject Management Institute [7]. Figure 2 presentsthe steps for risk management that are proposed bythe PMI.It is a self-proving fact that Fig.1 and Fig.2 are ingeneral the same process where the product orproject manager are assigned the responsibility of  n   investigating and revealing all the potential risks n   evaluate their importance (both methods useprobability of appearance and potentialconsequences) n   decide what are the important risks that shouldbe dealt with first n   follow up the overall process 3   Problem Solution 3.1Common Risk Management Approach Based on the aforementioned observations, it can beconcluded that a common procedure for technicaland managerial risks may be applied. As earliermentioned, a basic factor for an efficientmanagement of risk is the corporate memory. TheRisk Management Corporate Memory (RMCM)supports corporate knowledge management of risk atbusiness level [1]. Risks are organised intocategories and an organisational risk management“referential” is constructed. The referential isconsisted of all the risks concerning productdevelopment and have been identified in theenterprise, structured as shown in Fig. 3 . Dependingon the available information, a risk can be describedin two different ways [6], a short and a long one.In Fig. 3 , each block represents an ontology relatedto risk. Each ontology is described by a specificcharacteristic, appearing in the elliptic scheme. Moreprecisely the definitions are described below(adaptation from [2]):  n   risk is a hazard or opportunity for the product.Risk severity is described by its exposure. TheExposure is calculated by taking into accountthe probability of the cause(s) and the gravity of impact(s) related to the risk (see Algorithm). n   cause is any uncertain event that may lead tothe appearance of the risk and it is described byits probability of occurrence. A risk might haveseveral causes. n   impact is the effect of the risk on the product orsystem and it is described by its gravity. A risk might have several impacts. n   Action is the way in which the product managercan cope with the risk and it is described by itscost. Actions are either preventive or corrective.Preventive actions are those who aim at theprobability of occurrence of the cause andcorrective those that aim at the gravity of theimpact. Risk ActionImpactCause Long descriptionShort description exposuregravitycostprobability Fig. 3: Risk structure The short way of describing a risk is to give just itsexposure, most of the times coming from experts’opinion. Possible actions to mitigate the risk are alsoreferenced.The long way of describing a risk is by taking intoaccount a more complete reasoning mode about therisk, called the complete causal chain. The completecausal chain is composed of risk, risk cause(s), risk impact(s) plus the action(s) to mitigate the risk. Allthe important information about the causal chain isstored in the RMCM in the form of a risk sheet [3]. 3.2Proposed Procedure The proposed procedure is divided into five basicsteps. The identification of risks, which is the firststep, is based on the risk referential. The referentialis reviewed and the risks that can be related to theproduct under investigation are selected. Moreover,new risks that are identified for the first time in thecurrent product are also taken into account andenhance thereafter the referential.The product manager is entirely in charge of the risk identification. The referential is used as an auxiliaryhelp and the product manager is authorised to adaptit to his own context. In other words, the referentialaids the product manager to reuse the knowledgeexisting in the referential. Identification of risksassociated to a specific product can be made inseveral ways (by list, category or product element),depending on the level of existing knowledge. Theexisting Knowledge is organised in two dimensions.The first one concerns the amount of information,which is the level of detail for a risk (short or longdescription). The second dimension concerns theorganisation of Knowledge. There are three levelsof organisation. The first one, where there is only anunstructured list of risks, the second, where the risksare classified into categories such as managerial,financial, etc. and the last and most meaningful,where risks are attached to a specific productelement such as a specific technical item or afunction, etc.The second step of the procedure is the assessment,which is based on a scale of risk. Scaleparameterisation should be specific to eachcompany. For instance, the highest level of risk could be estimated in human deaths in one case ormillions of Euros in another. In the model presentedin this paper, the scale is divided into two majorsub-scales in order to be as much adaptable tospecific industrial needs as possible. The first sub-scale is quantitative, expressed in terms of moneyand the other one is qualitative, based on a specificindicator. The model covers two basic needs of acompany: n   It gives the possibility to describe the risk interms of money. This is a benefit for thecompany as the risk could be compared to ROImeasures. Moreover, bank, loans and financing,needs terms of money to proceed. n   On the other hand, most companies usequalitative systems because it is faster andeasier to produce and it does not require sodetailed information. However, it lucksprocessing capabilities and it boxes up lessinformation.Risks are considered either as hazards oropportunities. The cumulative effects of uncertainoccurrences, which affect the product or the relatedsystem positively, are considered as opportunities  (positive risks). Events with potential for harmfulconsequences to the design, operation orenvironment of the product or system areconsidered as hazards (negative risks).Knowledge state and knowledge progress aboutvalues of risks (i.e. first estimation at the beginningof the design, second after the implementation of amitigation action, third return from experience, etc)are managed in the method presented, within threestates which describe cause’s, impact’s and risk’sassessment: n   The estimated value is the evaluation that existsin the referential at the beginning of the designprocess concerning the causal chain element inregard to the specific product. n   The initial value – before any mitigation actionhas been implemented is the product manager’sopinion at the beginning of the design process,after taking into account the referential aboutthe specific causal element. n   The reduced value is the estimation for thecausal chain element after the implementationof mitigation action(s). The product managerfills in this value during the design ordevelopment process.When the evaluation of the identified risks iscompleted a global exposure for each product iscalculated. In that way, each product ends up withtwo indexes. The first is the cumulative result of themoney scale (overall calculations can be done onlywhen numeric information exists) and the second isa list of the qualitative assessed risks prioritised bytheir linguistic exposure (high, medium, low).The manager in charge, based on these two indexescan make the comparison of the existing productsolutions, if there are more than one.The third step is risk prioritisation, which is judgedfrom their exposure. The prioritisation is adescending sorting of exposures. When thequantitative scale (exposure in money) is used, theProduct Manager is responsible for determining anexposure threshold alarm. This threshold is used toexpress the limit above which risks are regarded asimportant.When the qualitative scale is used, three differentresults for the exposure are given. Prioritisation isan order of “unacceptable”(high), “to beexamined”(medium) or “non important” (low) risks.Risk mitigation, which is the fourth step of theprocedure, is the way that has to be followed inorder to minimise negative risks and maximise thepositive ones. The important risks (unacceptable –to be examined) that occurred after the assessmentstep should be closely examined and treated, bydefining the appropriate mitigation actions. Anaction can have affect on a cause of a risk – byreducing (or increasing for opportunities) itsprobability of occurrence - or on an impact – byreducing (or increasing for opportunities) itsgravity. Product Manager has the responsibility toevaluate the possible actions and judge for theireffectiveness in regard to their cost.The last step is Risk follow-up during the design ordevelopment phase. There are many cases whererisk exposure changes during the evolution of thedesign or development of a product. Aim of theFollow Up procedure is the updating of informationconcerning the risk causal chain (risk, cause, impactand actions), as well as the monitoring of theefficiency of mitigation actions. 3.3The algorithm The scope of this part of the paper is to describe thealgorithm used for the implementation of the modelconcerning the product risk estimation. As statedearlier, the assessment of risk and consequently thealgorithm is distinguished into quantitative andqualitative forms, as described below.As far as the quantitative assessment is concerned,the method provides the calculation of the estimated,initial and reduced exposure of each risk. Eachimpact and risk is expressed quantitatively in termsof money.The exposure of a risk  (estimated, initial or reduced) is calculated by the multiplication of “causeprobability” times “impact gravity”. In case aspecific risk is linked with more than one cause thenthe “cause probability” is calculated by using theappropriate probabilities laws [6]. As far as thisalgorithm is concerned, the “cause probability” iscalculated by the eqn (1):  n P = P 1 ∪ P 2 ∪ …P n = 1 - ∏ (1-P i ) (1) i=1 Where, P i is the probability of occurrence of the i-thcause, for i=1,2,..,nand 0 ≤ P i ≤ 1.For n=2 the eqn (1) reduces toP 1 ∪ P 2 = P 1 + P 2 - P 1 P 2  (2)In case a specific risk is linked with more than oneimpact, the “impact gravity” is the sum gravity of the impacts linked to the risk, as follows:  n G = ∑ G i (3)  i=1 Where, G i is the gravity of the i-th impact, fori=1,…,n  Thus, in general each type of exposure is beingcalculated by using the following equations: Estimated exposure: E E = P E *G E (4)Where, P E : cause estimated probability (using eqn(1)), G E : impact estimated gravity (using eqn (3)) Initial exposure: E I = P I *G I   (5)Where, P I : cause initial probability (using eqn (1)),G I : impact initial gravity (using eqn (3)) Reduced exposure: E R = P R *G R (6)Where, P R : cause reduced probability (using eqn(1)), G R : impact reduced gravity (using eqn (3))Concerning the calculation of the global exposuresfor a specific product the following equations areused: n Total Initial Exposure: TIE = ∑ E Ii (7)  i=1 Where,E Ii : the initial exposure of the i- th risk linked to aspecific proposal   , for i=1,…,n n Total Reduced Exposure: TRE = ∑ E Ri (8)  i=1 Where, E Ri : the reduced exposure of the i- th risk linked to a specific proposal   , for i=1,…,nAs far as the qualitative assessment is concerned, a5-scale measurement (suggestively: very high, high,medium, low and very low) has been chosen to beused. The qualitative scale will be used if theproduct manager is not able or does not want toestimate the risk in terms of money. Each risk exposure will be assigned according to the matrix of  Fig.4 , where, the probability of the cause and thegravity of the impact will be expressed by one of thelinguistic values (Very low, Low, Medium, High,Very high). 4 Conclusion – Further research This paper presented an integrated model coveringthe needs of product risk estimation. The two mostwidely applied methodologies of risk managementhave been reviewed and it has been shown that theyfollow a very similar approach. The concept of thedevelopment of a Risk Management CorporateMemory has been proposed and the model wascompleted with an algorithm, able to provide anindex of the overall risk exposure. Based on thisindex, the Product Manager may decide which is thebest solution among different possible products orwhether the risk exposure is acceptable or not for aspecific product.Further research should take place in the field of Knowledge Management in order the appropriatetechniques of gathering information for the Risk Management Corporate Memory to be revealed.  References: [1] Alquier A.M., Salles M., Tignol M.H., ProjectManagement Technique to Estimate and ManageRisk of Innovative Projects,  Nordnet’01 ,Stockholm, 2001.[2] Bellos V., Kirytopoulos K., Leopoulos V.,Sfantsikopoulos M., Innovative method for datacollection and risk analysis during the biddingprocess,  Risk Analysis 2002 Conference , Sintra,Portugal, 2002.[3] Chapman C, Ward S., Project risk management:Processes, Techniques and Insights , John Wiley& Sons Ltd, UK, Sussex, 1997.[4] Dhillon S.,  Design Reliability: Fundamentalsand Applications , CRC Press, 1999.[5] MIL-F-18372 (Aer), General Specification forDesign, Installation, and Test of Aircraft FlightControl Systems, Bureau of Naval Weapons,Department of the Navy, Washington.[6] PRIMA Consortium, “Description of thePRIMA method”, European Project IST- 1999 –10193, 2001.[7] Project Management Institute, Project management book of knowledge, PMI, 1996. ProbabilityExposure Very highMEDIUMMEDIUMHIGHHIGHHIGH   HighLOWMEDIUMMEDIUMHIGHHIGHMediumLOWMEDIUMMEDIUMMEDIUMHIGHLowLOWLOWMEDIUMMEDIUMMEDIUMVery lowLOWLOWLOWLOWMEDIUMvery lowlowmediumhighVery high Gravity Fig. 4: Risk Matrix
Similar documents
View more...
Related Search
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks