An approach to risk Assessment for Outsourcing

Download PDF
34 downloads 0 Views 753KB Size Report
Comment
the stakeholders to answer the question whether outsourcing is feasible for their case ..... FMEA uses ... required by the risk assessment, multiple factors may be.
Management and on Control IFAC Conference Manufacturing Modelling, IFAC Conference on Manufacturing Modelling, June 28-30, 2016. Troyes, France Modelling, Management and on Control IFAC Conference Manufacturing Management and Control Available online at www.sciencedirect.com June 28-30, 2016. Troyes, France Management and Control June 28-30, 2016. Troyes, France June 28-30, 2016. Troyes, France

ScienceDirect

IFAC-PapersOnLine 49-12 (2016) An approach to risk Assessment for 1239–1244 Outsourcing Logistics: Case of An approach to risk Assessment for Outsourcing Case of An for Outsourcing Logistics: Case of Pharmaceutical Industry Logistics: An approach approach to to risk risk Assessment Assessment for Outsourcing Logistics: Case of Pharmaceutical Industry Pharmaceutical Industry Pharmaceutical IndustryA. El Mhamedi* A. El Mokrini*, E. Dafaoui*, A. Berrado**,

A. A. El El Mokrini*, Mokrini*, E. E. Dafaoui*, Dafaoui*, A. A. Berrado**, Berrado**, A. A. El El Mhamedi* Mhamedi* A. de El Paris8 Mokrini*, E. Dafaoui*, A. de Berrado**, El Mhamedi* *MGSI – Université / LISMMA, 140, rue la nouvelleA.France, 93100 Montreuil, France *MGSI –– Université de 140, France, [email protected]) (e-mail: [email protected], [email protected], *MGSI Université de Paris8 Paris8 // LISMMA, LISMMA, 140, rue rue de de la la nouvelle nouvelle France, 93100 93100 Montreuil, Montreuil, France France *MGSI – Université de Paris8 / LISMMA, 140, rue de la nouvelle France, 93100 Montreuil, France (e-mail: [email protected], [email protected], [email protected]) **AMIPS-Ecole Mohammadia d’Ingénieurs, Université Mohamed 5 Avenue Ibn Sina B.P. 765 Agdal, Maroc (e-mail: [email protected], [email protected], [email protected]) [email protected]) (e-mail: [email protected], [email protected], **AMIPS-Ecole Mohammadia d’Ingénieurs, Université Mohamed 5 Avenue Ibn Sina B.P. 765 Agdal, [email protected] ) (e-mail : **AMIPS-Ecole Mohammadia d’Ingénieurs, Université Mohamed 5 Avenue Ibn Sina B.P. 765 Agdal, Maroc Maroc **AMIPS-Ecole Mohammadia d’Ingénieurs, Université Mohamed 5 Avenue Ibn Sina B.P. 765 Agdal, Maroc (e-mail [email protected])) (e-mail :: [email protected] (e-mail : [email protected]) Abstract: Outsourcing functions such as logistics has become an industry trend towards costAbstract: Outsourcing functions such as as logistics logistics has become an industry industry the trend towards costeffectiveness and high service level performance. Many has firmsbecome have acknowledged benefits of relying Abstract: Outsourcing functions such an trend towards costAbstract: Outsourcing functions such astheir logistics has become an industry trend towards costeffectiveness and high service level performance. Many firms have acknowledged the benefits of relying on external experts in a need to empower abilities. The pharmaceutical industry in particular is effectiveness and high service level performance. Many firms have acknowledged the benefits of relying effectiveness andconstant high service level performance. Many firms have acknowledged the benefits of relying on external experts in a need to empower their abilities. The pharmaceutical industry in particular is challenged by evolution of their development and manufacturing processes. Outsourcing on external experts in a need to empower their abilities. The pharmaceutical industry in particular is on external experts in aanneed to empower their abilities. The pharmaceutical industry in Outsourcing particular is challenged by constant evolution of their development and manufacturing processes. logistics becomes then attractive option for firms to focus on their core competencies. However, challenged by constant evolution of their development and manufacturing processes. Outsourcing challenged by numerous constant their for development and manufacturing processes. Outsourcing logistics becomes becomes then an anevolution attractive option firms to focus focus on their core competencies. However, alongside the benefits ofofoutsourcing, various risks arise with thecompetencies. implementation of this logistics then attractive option for firms to on their core However, logistics becomes then an attractive option for firms to focus on their core competencies. However, alongside the numerous benefits of outsourcing, various risks arise with the implementation of into this option. contributes literature various by presenting a decision that takes alongsideThis the paper numerous benefits to of the outsourcing, risks arise with the model implementation of this alongside the numerous benefits of outsourcing, various risks arise with the implementation of this option. This paper contributes to the literature by presenting a decision model that takes into consideration outsourcing to logistics in the pharmaceutical chain. Risk identification option. This risks paperofcontributes the literature by presentingsupply a decision model that takesand intoa option. This risk paper to the literature by a decision model that takesand into consideration risks ofcontributes outsourcing logistics in the the pharmaceutical pharmaceutical supply chain. Risk Risk identification multi-criteria assessment model using ELECTRE TRIpresenting are described. consideration risks of outsourcing logistics in supply chain. identification and aa consideration risksassessment of outsourcing logistics in the pharmaceutical supply chain. Risk identification and a multi-criteria risk model using ELECTRE TRI are described. multi-criteria risk assessmentFederation model using ELECTRE TRI areHosting described. © 2016, IFAC (International of Automatic by Elsevier Ltd.(PSC) All rights reserved. Keywords: Risk Assessment, Logistics,Control) Pharmaceutical supply chain multi-criteria risk assessment Outsourcing, model using ELECTRE TRI are described. Keywords: Risk Assessment, Outsourcing, Logistics, Pharmaceutical supply chain (PSC) Keywords: Risk Assessment, Outsourcing, Logistics, Pharmaceutical supply chain (PSC)  Pharmaceutical supply chain (PSC) Keywords: Risk Assessment, Outsourcing, Logistics,  happen in the future. We can use the best forecasts and do  1. INTRODUCTION  happen in use best forecasts do every possible analysis,We butcan there always about happen in the the future. future. We can useisthe the best uncertainty forecasts and and do 1. INTRODUCTION 1. INTRODUCTION happen in the future. We can use the best forecasts and do every possible analysis, but there is always uncertainty about future events. It is this uncertainty that brings risks. In order to survive 1. in INTRODUCTION a highly competitive and demanding every possible analysis, but there is always uncertainty about every possible analysis, butis there isthat always uncertainty about future events. It is this uncertainty brings risks. In order to survive in a highly competitive and demanding The purpose of this study to present a decision framework industry market, firms are nowadays required to constantly In order to survive in a highly competitive and demanding future events. It is this uncertainty that brings risks. futureprovide events. aIt this is this uncertainty that brings risks.framework In order totheir survive in a highly competitive andto demanding The study is aa for decision industry firms nowadays required risk assessment model considering the improve companies are and The purpose purpose of of this study is to to present present decision framework industry market, market, supply firms are arechains. nowadaysMany required to constantly constantly The purpose of this study is to present a for decision framework industry market, firms are nowadays required to constantly and provide a risk assessment model considering improve their supply chains. Many companies are option of logistics outsourcing. The remainder of this challenged to increase their levels of service performance the improve their supply chains. Many companies are and provide a risk assessment model for considering the providelogistics a is risk assessment model fortheconsidering the improve their supply chains. Many companies are and option outsourcing. The of challenged to increase their levels of performance communication organized as follows. next section, while keeping the supply chain cost-effective. a strategic option of of logistics outsourcing. The Inremainder remainder of this thisa challenged to increase their levels of service serviceOn performance option of logistics outsourcing. The remainder of this challenged to increase their levels of service performance communication is organized as follows. In the next section, while the supply On of literature about outsourcing is described. level, firms have morecost-effective. than ever interested in the review communication is organized as follows.logistics In the next section, aa while keeping keeping the become supply chain chain cost-effective. On aa strategic strategic communication is organized as follows. In the next section,ina while keeping the supply chain cost-effective. On a strategic review of literature about outsourcing logistics is described. level, firms have become more than ever interested in the We highlight the use of multi-criteria decisionisanalysis option of outsourcing to cope with competition. Outsourcing of literature about outsourcing logistics described. level, firms have become more than ever interested in the review review of literature about outsourcing logistics is described. level, firms have become more than ever interested in the We highlight the use of multi-criteria decision analysis in option of outsourcing to cope with competition. Outsourcing outsourcing general andmulti-criteria in risk assessment particular. functions that are generally the field of Outsourcing expertise of We highlightinthe use of decisionin analysis in option of outsourcing to copeoutside with competition. We highlight the use of multi-criteria decision analysis in option of outsourcing to cope with competition. Outsourcing outsourcing in general and in risk assessment in particular. functions that are generally outside the field of expertise of We then setinthe context work. firms, suchthat as logistics, allows organizations on their general andand in the risk motivation assessment of in this particular. functions are generally outside the fieldtooffocus expertise of outsourcing inthe general and in the risk motivation assessment in assessment particular. functions that are generally outside the field offocus expertise of outsourcing We then set context and of this work. firms, such as logistics, allows organizations to on their Section 3 deals with risk identification stages and core competencies while these functions are performed by an firms, such as logistics, allows organizations to focus on their We then set the context and the motivation of this work. then set thewith context and the motivation of this work. firms, such as logistics, allows to focus onby their Section risk stages and core while functions are methods. We by presenting an ELECTRE TRI external service provider. Theorganizations pharmaceutical industry in We Section 3 3 deals dealswill withfollow risk identification identification stages and assessment assessment core competencies competencies while these these functions are performed performed by an an Section 3 deals withfollow risk identification stages and assessment core competencies while these functions are performed by an methods. We will by presenting an ELECTRE external service provider. The pharmaceutical industry in model for risk assessment in outsourcing. Finally, a summary particular is characterized by highly technical core functions TRI external service provider. The pharmaceutical industry in methods. We will follow by presenting an ELECTRE TRI We will follow by presenting an ELECTRE TRI external service provider.by The pharmaceutical industry in methods. for in Finally, aa summary particular is highly technical functions and of possible extensions conclude the that include development and manufacturing. Outsourcing modelpresentation for risk risk assessment assessment in outsourcing. outsourcing. Finally, summary particular is characterized characterized by highly technical core core functions model model for risk assessment in outsourcing. Finally, a summary particular is characterized by highly technical core functions and presentation of possible extensions conclude that include development and manufacturing. Outsourcing communication. logistics for this industry allows firms to counter challenges the that include development and manufacturing. Outsourcing and presentation of possible extensions conclude the and presentation of possible extensions conclude the that include development and manufacturing. Outsourcing logistics industry allows to challenges related tofor constant evolution of medicinal development communication. 2. LITERATURE REVIEW logistics forthethis this industry allows firms firms to counter counter challenges communication. logistics for this industry allows firms to counter challenges related to evolution medicinal and manufacturing. However, industry communication. 2. LITERATURE REVIEW related to the the constant constant evolutiontheof of pharmaceutical medicinal development development 2. LITERATURE related to the constant evolution of medicinal development and manufacturing. However, the pharmaceutical industry has specificities that However, distinguishthe it from other domains such Logistics outsourcing as defined byREVIEW Lynch et al. (2006) is an and manufacturing. pharmaceutical industry 2. LITERATURE REVIEW and manufacturing. However, the pharmaceutical industry has specificities that distinguish it from other domains such Logistics outsourcing Lynch et is as product perishability, crucial importance of safety and arrangement whereby logisticsby performs has specificities that distinguish it from other domains such Logistics outsourcing as asa defined defined byservice Lynchprovider et al. al. (2006) (2006) is an an has specificities thatand distinguish itimportance from otherguidelines, domains such Logisticsfor outsourcing asathat defined bybe, Lynch et al. (2006) is an as product perishability, crucial of safety and arrangement whereby logistics service provider performs security, legislation jurisdictional quality services a company could or have been, provided as product perishability, crucial importance of safety and arrangement whereby a logistics service provider performs as product perishability, crucial importance of safetytypes and arrangement logistics servicehave provider performs security, legislation and jurisdictional quality aa whereby company could be, been, special storing and distributing different in-house. The literatureathat allows classifying outsourcing work security,conditions legislationfor and jurisdictional quality guidelines, guidelines, and services services for for company that could be, or or have been, provided provided security, legislation and jurisdictional quality guidelines, and services for a company that could be, or have been, provided special conditions for storing and distributing different types in-house. The literature allows classifying outsourcing of products. Due to these specificities, the feasibility of done by researchers to 5 main categories starting with process special conditions for storing and distributing different types in-house. The literature allows classifying outsourcing work work special conditions for storing and different in-house. The literature allowscategories outsourcing work of Due the of done to 5 starting with process outsourcing needs specificities, to distributing be investigated, andtypes an frameworks etclassifying al., 2006), risks associated of products. products. logistics Due to to these these specificities, the feasibility feasibility of decision done by by researchers researchers to(Kremic 5 main main categories starting with process of products. Due to these specificities, the feasibility of done by researchers to 5 main categories starting with process outsourcing logistics needs to be investigated, and an frameworks (Kremic al., 2006), associated appropriate decision must chosen. Firms with outsourcing (Schniederjans what outsourcing logistics process needs to be beinvestigated, and are an decision decision frameworks (Kremic et etand al.,Zuckweiler, 2006), risks risks2004), associated outsourcing logistics needs to be of investigated, and an with frameworks (Kremic al.,2002), 2006),torisks associated appropriate decision process must Firms are outsourcing (Schniederjans Zuckweiler, 2004), what nowadays aware of the advantages outsourcing which activities to outsource (Platts etetand al., what extent to appropriate decision process must be be chosen. chosen. Firms are decision with outsourcing (Schniederjans and Zuckweiler, 2004), what appropriate decision process must be chosen. Firms are with outsourcing (Schniederjans and Zuckweiler, 2004), what nowadays aware of the advantages of outsourcing which activities to outsource (Platts et al., 2002), to what extent include reducing costs; exchanging fixed costswhich with outsource et al., 2007), and2002), results outsourcing nowadays awareoperating of the advantages of outsourcing activities to(Kotabe outsource (Platts et al., to of what extent to to nowadays awareelimination of the advantages of outsourcing which outsource activities to(Kotabe outsource (Platts et2004). al., 2002), to of what the extentfirst to include operating fixed costs et variable costs; of exchanging infrastructure investments; (Berggren and Bengtsson, Regarding include reducing reducing operating costs; costs; exchanging fixed costs with with outsource (Kotabe et al., al., 2007), 2007), and and results results of outsourcing outsourcing include reducing operating costs; exchanging fixed costs with outsource (Kotabe et al., 2007), and results of outsourcing variable costs; elimination of infrastructure investments; (Berggren and Bengtsson, 2004). Regarding the first improved ability to sense and respond real-time to changes category, many authors worked on providing decision variable costs; elimination of infrastructure investments; (Berggren and Bengtsson, 2004). Regarding the first variable costs; elimination of risk; infrastructure and Bengtsson, 2004). Regarding the first improved ability and changes category, authors worked on providing decision business sharing andreal-time access investments; toto frameworks to guide decision makers the process of improvedenvironments; ability to to sense sense and respond respond real-time toresources changes (Berggren category, many many authors worked onduring providing decision improved ability to sense and respond real-time toHowever, changes category, many authors worked on providing decision business environments; sharing risk; and access to resources frameworks to guide decision makers during the process not available in-house (Persson and Virum, 2001). outsourcing (Miranda and Kim, 2006). This first phase allows business environments; sharing risk; and access to resources frameworks to guide decision makers during the process of of business sharing risk;Virum, andtoaccess to However, resources to guide decision during theoutsourcing process of not in-house and 2001). outsourcing (Miranda and 2006). first there are environments; a number risks that be identified and frameworks the stakeholders to answer the makers question whether not available available in-houseof(Persson (Persson andneed Virum, 2001). However, outsourcing (Miranda and Kim, Kim, 2006). This This first phase phase allows allows not available in-house (Persson and Virum, 2001). However, (Miranda and Kim, 2006). first phase in allows there are number of risks to and the stakeholders answer the whether evaluated going process outsourcing. is for to their case not. This Outsourcing the there are aawhen number of through risks that thattheneed need to be beofidentified identified and outsourcing the feasible stakeholders to answer theorquestion question whether outsourcing outsourcing there are a number of risks that need to be identified and the stakeholders to answer the question whether outsourcing evaluated when going through the process of outsourcing. is feasible for their case or not. Outsourcing in Risks occur because can never what will is pharmaceutical in particular been investigated on evaluated when goingwethrough the know processexactly of outsourcing. feasible forindustry their case or not.has Outsourcing in the the evaluated when going through the processexactly of outsourcing. feasible for their case or not. Outsourcing in the Risks pharmaceutical Risks occur occur because because we we can can never never know know exactly what what will will is pharmaceutical industry industry in in particular particular has has been been investigated investigated on on Risks occur because we can never know exactly what will pharmaceutical industry in particular has been investigated on

Copyright © 2016 IFAC 1239 2405-8963 © IFAC (International Federation of Automatic Control) Copyright © 2016, 2016 IFAC 1239Hosting by Elsevier Ltd. All rights reserved. Copyright 2016 responsibility IFAC 1239Control. Peer review©under of International Federation of Automatic Copyright © 2016 IFAC 1239 10.1016/j.ifacol.2016.07.681

IFAC MIM 2016 1240 June 28-30, 2016. Troyes, France

A. El Mokrini et al. / IFAC-PapersOnLine 49-12 (2016) 1239–1244

many levels of the supply chain. On the strategic level, (USAID, 2010) proposed a decision framework to start the process of outsourcing in the case of the pharmaceutical industry. In a 2012 report, the OMS (Organisation Mondiale de la Santé) highlights the importance of providing a clear perspective to assess the feasibility of vaccines logistics outsourcing from the public to the private sector. In another work, An important phase in decision the process of outsourcing deals with the identification and assessment of potential risks. Risk as defined by Zsidisin and Ritchie (2008) are “the probability of an incident associated with inbound supply from individual supplier failure or the supply market occurring, in which its outcomes result in the inability of the purchasing firm to meet customer demand or cause threats to customer life and safety". The first step in risk assessment is related to the identification of potential risks. Several classifications of risks have emerged in literature. For example, Harland et al. (2003) classified them as operational, strategic. When dealing with risk identification in the pharmaceutical industry, specific characteristics need to be taken into consideration. Enyinda et al. (2009) propose to classify (PSC) outsourcing risks to regulatory risk, operational risk, technical risk, and corporate social responsibility risk. They state that supply chain risk management strategies that can be employed to tame a firm’s exposure to supply chain outsourcing risk are risk avoidance, reduction, transfer, and acceptance. Other research in the field of pharmaceuticals presented investigations about risk management strategies in general (Woitach, 2007), and particularly for the cold chain (Bishara, 2006). Another phase of outsourcing deals with the selection of service provider. Numerous authors presented methods for selecting the most appropriate service provider based on different criteria, we site Bansal and Kumar (2013). Methods of service provider evaluation range from simple analytical techniques to multi-criteria methods. However, multi-criteria decision analysis (MCDA) methods have not only been applied to service provider selection phase of outsourcing; researchers have used the techniques at different phases involved in the decision process. For example, Hahn et al. (2014) tackled the issue of contract manufacturing at the strategic-tactical level using a multi-criteria decision- making perspective, since service, cost, quality, and more long-term value-related aspects are involved. They applied a scenariobased approach that is built within the concept of Data Envelopment Analysis. Hassanain et al. (2015) propose the use of the Analytic Hierarchy Process (AHP), as a multicriteria decision-making model, for use by maintenance managers to consider before making a decision on outsourcing. MCDA techniques have been used in many phases of outsourcing; however, few studies pertained to their application to the risk assessment phase of outsourcing. In fact, stakeholder preferences for risk assessment indicators and assessment deliverables hold great importance but are often neglected (Scheuer et al., 2013). MCDA actually provides structured methods for the incorporation of project stakeholders' opinions in the ranking of alternatives (Linkov et al., 2006). For example, Kull and Talluri (2008) consider extant research in supply risk in developing a framework for

risk assessment based on various categories and types of risks. They propose a combination of AHP and goal programming as a decision tool for supplier selection in the presence of risk measures and product life cycle considerations. Brito et al. (2010) propose a multi-criteria model for assessing risk in natural gas pipelines, and for classifying sections of pipeline into risk categories. The model integrates Utility Theory and the ELECTRE TRI method. Scheuer et al. (2013) also applied MCDA to flood risk management which possibly reflects the individual characteristics of all elements at risk of being flooded. Pertaining to the pharmaceutical supply chain, few studies incorporate MCDA in risk assessment. For example, Enyinda et al. (2009) investigated risk management in pharmaceutical global supply chain outsourcing using the MCDA technique AHP. In another work, Jaberidoost et al. (2015) studied risk assessment in pharmaceutical industry in Iran through a questionnaire and consultation with experts using group AHP method as well with the support of a rating scale and a risk evaluation of simple additive weighting method. MCDA methods have been widely applied in risk management; however, they have usually been used for risk ranking as opposed to risk evaluation or level determination. 2. CONTEXTUAL SETTING The motivation of this work is drawn from the fact that logistics outsourcing has become an attractive option for pharmaceutical firms as a way to increase levels of service performance. These firms are asked today to react to fluctuating demands in a highly responsive and flexible way. However, increasing service performance usually goes hand in hand with increasing costs which presents a conflicting situation for decision makers. Logistics outsourcing service providers have attained high levels of expertise and optimization today, which creates a competitive market presenting interesting offers to different industry domains. Outsourcing logistics for pharmaceutical firms presents the ability to focus on their technical expertise with the possibility of reducing costs and increasing levels of service performance. However, in order to consider outsourcing logistics, it is crucial to assess the feasibility of this option. Numerous factors should be taken into consideration whilst assessing the feasibility of the outsourcing for the pharmaceutical industry. The figure below (Fig.1) presents the main influencing factors and their attributes (El Mokrini et al., 2015). Strategy and market investigation constitute the first two influencing factors. Strategy comprises three attributes starting with the perspective which is related to either the need for cost savings, need for resources, focus on core competency, or value creation. Once strategy and market attributes have been defined, the next step is to investigate cost savings that represent the financial factor. Followed by the service performance factor which requires identifying the needed levels of reliability, responsiveness, flexibility and quality. The next influencing factor deals with the technological capabilities that the firm needs to attain its objectives.

1240

IFAC MIM 2016 June 28-30, 2016. Troyes, France

A. El Mokrini et al. / IFAC-PapersOnLine 49-12 (2016) 1239–1244

1241

consequences as input criteria for the MCDA method selected. The result of this analysis will be incorporated in an ELECTRE TRI model formulation that will take into account the decision maker’s judgment to assign the identified risks to categories of risk levels: Very High, High, Medium, Low, and Remote. The following sub-sections will present a description of risk assessment processes.

Fig.1 Factors influencing logistics outsourcing Relating to the pharmaceutical industry, the pharmaceutical products supply chain (PPSC) specificities factor requires determining the requirements of storage sites and transportation means. Finally, risk assessment constitutes an important step in the process of assessing the feasibility of the outsourcing option. These risks are related to all factors stated before and they influence considerably the decision of outsourcing. For this reason, a proper investigation on risk assessment is in order when dealing with a strategic decision such as outsourcing. As an extension to Fig.1, later sections will elaborate risk assessment by detailing the processes of risk identification and evaluation. 3. PROPOSED APPROACH Risks need to be given a thorough assessment as an important dimension of outsourcing logistics. This work extends the decision framework presented above to describe the risk factor and include an innovative technique for risk evaluation. This study has 4 output processes: identification of the most relevant risks related to outsourcing in the pharmaceutical supply chain, determination of criteria for the evaluation of those risks, selection of an appropriate MCDA method, and finally presentation of a model formulation for the problem. These outputs along with their input processes are presented in Fig. 2. As a first step, risks related to outsourcing logistics need to be identified. The consequent evaluation step could be undertaken using many already-set methods; however, these methods do not take into consideration the preferences of the decision maker. In this paper, we are interested in using MCDA methods in order to incorporate the decision maker priorities and concerns into risk assessment. These methods have been previously applied for risk prioritization; however, very few researches have used MCDA sorting techniques for risk evaluation. This paper’s contribution lies in first identifying and classifying relevant risks into categories and defining the risk evaluation problem from a different perspective. This is obtained by the use of risk components such as probability of occurrence and

Fig.2 Methodology input and output processes 3.1 Risk Assessment As part of the factors influencing the outsourcing logistics decision in the pharmaceutical industry, risks represent an important factor that requires first-hand assessment. Assessing risks when making a decision like outsourcing involves a great deal of uncertainty. The following figure (fig 3) presents the different steps related to risk management. They are split into 2 categories: risk assessment and risk control. Risk assessment contains three processes: risk identification, risk analysis, and risk evaluation. Each stage of the assessment process will be described in the following subsections. However, the control process which involves risk mitigation and monitoring is not included at this stage of the decision making since the aim is to assess generally the risks to determine whether they are acceptable or not.

Fig.3 Risk assessment processes 3.2 Risk Identification

1241

IFAC MIM 2016 1242 June 28-30, 2016. Troyes, France

A. El Mokrini et al. / IFAC-PapersOnLine 49-12 (2016) 1239–1244

The first sub-process in risk assessment is risk identification which includes finding, distinguishing, and defining the risks and their categories that are involved with the logistics outsourcing. Based on literature investigation, we identified 6 mostly relevant categories of risks based on the needs for logistics outsourcing the pharmaceutical supply chain. Operational risks include risks related to the processes of storage and transportation, the other general categories include financial, technology-related, information related, and relational risks. The following Table.1 summarizes the most relevant risks encountered with outsourcing logistics. Table 1. Categories of risks in Outsourcing Logistics Category Operational: Storage and distribution processes

Risk R1- Poor infrastructure and handling R2- Delivery delay R3- Poor service performance: failure in attaining an acceptable degree of responsiveness and flexibility R4- Disorder of operations at the transition phase R5- Medicinal products quality deterioration

Financial

R6-Unrealized savings with possible increased costs R7-Loss of customers due to the partner’s poor service performance R8-Poor information system integration leading to failure in product traceability R9-Strategic and operational information leakage R10-Latent information asymmetry between client and service provider. (Wang and Regan 2002) R11-Poor supplier relations: Poor morale/employee issues R12- Reduced customer/supplier contact R13- Conflicts of culture R14- Poor contract (unclear) or poor planning of outsourcing functions R15- Poor selection of partner R16- Unrealistic or high expectations regarding the service provide performance R17- Lack of skills to evaluate, manage and monitor outsourced functions R18- Loss of control over the service provider

Technology Information Related Relational

Internal

Table 2. Risk analysis methods Assessment Method the failure modes and effect analysis Simple additive weighting (SAW) Quality function deployment (QFD) Stochastic Dominance Consequence/probability matrix Fault tree analysis

References KUMARA et al (2014) JABERIDOOST et al. (2015) WANG et al. (2014) CHANG et al. (2015) ALE et al. (2015) MAHMOOD et al. (2013)

3.4 Risk evaluation Risk evaluation presents the third sub-process of the assessment process. It uses the output of the analysis –level of the risk- and the priority of this risk and compare it with the thresholds of acceptance. If the risk is not acceptable then the next step is to investigate the availability of strategies to reduce or eliminate it. As stated before, we are interested in the use of MCDA methods to evaluate risks. There are two possibilities to incorporate the decision maker’s preferences in risk evaluation. The first is ranking the risks into a priority list. This process can be done using AHP, Promethee or other ranking MCDA techniques. The second option is to sort the risk into categories. 3.4.1 Criteria used in risk classification

3.3 Risk Analysis The second sub-process in risk assessment is the risk analysis which is involved first defining the causes and consequences of the risks. The International Organization for Standardization (ISO, 2009) defines two of the essential components of risk: losses (quantified) and uncertainty of occurrence. Manuj and Mentzer (2008) present 3 questions to answer for risks: 1.What are the potential losses, if the risk is realized, what losses would result? 2. What is the probability of occurrence of these losses: the likelihood (probability) of the occurrence of an event that may lead to the realization of the risk? 3. What is the importance of the consequences of the loss? Several models have been applied to answer these questions. Examples are the Failure mode and effects analysis (FMEA), Consequence/probability matrix, Multi-criteria decision methods, Fault tree analysis, and other. FMEA uses a step‐by‐step approach for identifying all possible failures in a design, a manufacturing or assembly process, or a product or service (Xiao et al., 2011). The following table summarizes risk analysis methods and literature examples:

Depending on the level of detail and degree of objectivity required by the risk assessment, multiple factors may be employed to evaluate risks. As stated earlier in this document, the parameters generally used to assess risks are likelihood/probability of occurrence and impact. Detectability is another parameter that has been added to assess the level of risks. In our work, we use these parameters as criteria in the application of MCDA techniques. The following list presents the criteria used in our framework: 1- Financial impact 2- Impact on the PSC reactivity 3- Impact on patients’ safety 4- Impact on public regulation and legitimacy 5- Impact on organisational culture 6- Likelihood of occurrence 7- Risk mitigation effort The criteria include different categories of impacts which are defined by the degree of consequences incurred by the risks. In addition to the likelihood of occurrence which is evaluated depending on the scale used to compare risks according to the decision maker’s judgment. Unlike other risk assessment techniques, this one takes into consideration the risk mitigation effort which describes the degree of effort needed to be performed to mitigate the risk. 3.4.2 MCDA for risk sorting: choice of ELECTRE TRI MCDA methods of classification and sorting are based on specific functional forms. The use of these models infers that the corresponding specific function or relation is either outranking relation, utility function, or discriminant function. The von Neumann and Morgenstern Utility Theory presents an axiomatic approach that can assess the decision maker’s behaviour with regard to the risk (Keeney and Raiffa, 1976).These latter enable the representing the complexity of the decision-maker’s preferences. We have identified few

1242

IFAC MIM 2016 June 28-30, 2016. Troyes, France

A. El Mokrini et al. / IFAC-PapersOnLine 49-12 (2016) 1239–1244

techniques that pertain to our case and can be used for risk sorting in the case of outsourcing logistics. ELECTRE Tri (Figueira et al., 2013) for example uses the outranking concept by building binary relationships among each alternative and the profiles that bound each category (Dias et al., 2002). PROAFTN belongs to the class of supervised learning algorithms and enables to determine the fuzzy indifference relations by generalizing the indices (concordance and discordance) used in the ELECTRE Tri method. Then, it assigns the fitting degree of the alternatives to the categories. Another interesting methods that can be used is the multi-Group hierarchical discrimination (M.H.DIS) method. This method employs a hierarchical discrimination procedure to determine the class in which the alternatives under consideration belong. It leads to the development of a set of additive utility functions, which are used to decide upon the classification of each alternative into a specific group (Zopounidis and Doumpos, 2000). There exists other methods that be used risk classification such as the decision tree sorting, PREFDIS, Rangu system, etc. In this work, we are interested in ELECTRE TRI method for risk sorting since it has proven its efficiency and effectiveness in many successful applications in risk management (Govindan and Jepsen (2015), Taillandier et al. (2011). In addition, the outranking relation presents an important contribution to the decision maker. 3.4.3 Proposed Model formulation using ELECTRE Tri We have identified 18 alternative risks that need to be evaluated. ELECTRE TRI is used in our problem to assign risks to pre-defined categories (Fig 5).

Fig. 5. Risk level assignment CB={Cp=1, Cp=2, ….Cp=5} is the set of the 5 categories of risk level: Very High, High, Medium, Low, and Remote. The assignment of a risk r results from the comparison of r with the profiles defining the limits of the categories. Since the ELECTRE methods require that all criteria are defined on a common level, the criteria are considered at the same level. …, Let F define the set of the indices of these criteria (F = {1, 2,..., m = 7}) having seven criteria mentioned above. The profiles {bh=1, bh=2, bh=3, bh=4 bh=5} define the limits of categories (see Fig. 6, where the profiles b5 and b0 correspond to the ideal and anti-ideal risk, respectively). ELECTRE TRI assigns risks to categories following two consecutive steps. The first one is related to the construction of an outranking relation S that characterizes how risks compare to the limits of categories. This outranking relation is found through a pairwise comparison of the alternatives based on each criterion, to obtain an outcome richer than the dominance relations, and less rich but more realistic than multiple attribute utility or value functions (Roy and

1243

Bouyssou, 1991). The second step is related to the exploitation of the relation S in order to assign each risk to a specific category.

Fig. 6. Illustration of the risk sorting problematic This method enables building the outranking relation S, which validates or invalidates the assertion rSbh whose meaning is “risk r is at least as good as profile bh”. Relation S uses an index σ(r, bh ) € [0, 1] that represents the degree of credibility of the assertion rSbh. Other variables that are used in this method’s fulfilment are the indifference and and which characterize preference thresholds the intra-criterion preferential information. These thresholds represent the imprecise nature of the evaluations. On one represents the largest difference hand, that preserves indifference between risk r and bh on , and represents the smallest difference criterion that matches a preference in favor of r on criterion . As the assignment of alternatives to categories does not result directly from the relation S, an exploitation phase is necessary. The solution is to remove the fuzziness of the relation S using λ as a cutting level such that λ € [0.5,1]. The assertion rSbh is considered to be valid if σ(r, bh) ≥ λ σ(r, bh) ≥ λ. The latter λ -cut defines the preference position between r and, bh: - r is indifferent to bh when : σ(r, bh) ≥ λ and σ(bh, r) ≥λ - r is preferred to bh when: σ(r, bh) ≥ λ and σ(bh , r) < λ - bh is preferred to r when: σ(r, bh) < λ and σ(bh , r) ≥ λ - r is incomparable to bh when: σ(r, bh) < λ and σ(bh , r)