Supplier evaluation in local public services: Application of a model of value for customer

ARTICLE IN PRESS Journal of Purchasing & Supply Management 15 (2009) 33–42

Contents lists available at ScienceDirect

Journal of Purchasing & Supply Management journal homepage: www.elsevier.com/locate/pursup

Supplier evaluation in local public services: Application of a model of value for customer$ Alessandro Ancarani  Department of Civil and Environmental Engineering, Section of Economic, Managerial, Valuation Engineering, University of Catania, Viale A. Doria, 6, 95125 Catania, Italy

a r t i c l e in fo

abstract

Article history: Received 18 June 2008 Received in revised form 7 August 2008 Accepted 23 September 2008

A methodology for evaluating the supplier’s performance in a standardised public service is presented with a theoretical application to the water service. It aims at supporting the managers in their efforts to improve the service provided. The basic concept is that the quality of service depends on the quality of the supplier’s performances. It is then possible to determine such performances in terms of technical and functional quality as linked with customer satisfaction. It can be expressed in terms of the prevalent value model for the users and the quality of interaction between customer and supplier is a function of the quality of service provided (Qs). An agent-based model can be defined for simulating the interaction among the actors. & 2008 Elsevier Ltd. All rights reserved.

Keywords: Public services Supplier evaluation Agent-based model Value for customer Customer satisfaction

1. Introduction The public sector has sought to solve in the last decades the tensions between more market-based provision and more value-added public from the provision of public services (deLeon and Denhart, 2000). In spite of many critiques of the move to the externalization of services, by the late 1990s (among the others Milward, 1996; Milward et al., 1993) there was a much greater understanding than ever before that the market had a major potential role in supplementing—even sometimes replacing the role of public sector in-house provision. With this understanding, however, there came the realization that the public sector had traditionally misunderstood the range of possible relationships between actors in the ‘market’ (Kettl, 1993; Boyne, 1998). With the dawning appreciation that market relationships were socially constructed in the procurement process, not simply a product of ‘market conditions’, the need became clear for a more pro-active role by public sector organisations in market management (Filkin, 1998) and greater attention to developing mutually rewarding relationships with external contractors, whether they were in the public, private or voluntary sectors (Erridge and Greer, 2002). Also, the success or failure of any alternative servicedelivery arrangement likely depends on how well governments

$ An earlier version of this paper has been presented with M.F.M. Bonanno at the Ipsera 2007 Conference in Bath, UK.  Tel.: +39 095 7382718; fax: +39 095 7382748. E-mail address: [email protected]

1478-4092/$ - see front matter & 2008 Elsevier Ltd. All rights reserved. doi:10.1016/j.pursup.2008.09.003

can manage the entire contract process. Effective contract management requires mitigating specific problems that can plague the contract process. These problems arise from the dissatisfaction with prior contracting experiences, transaction costs, characteristics of government’s structure and operation and characteristics of external environment (Brown and Potoski, 2003). In the public sector since the 1980s there has been an international trend to contestability and market approach in the delivery of public services, mainly with reference to local public services. It has been paralleled by the adoption of contract instead of regulator contract (Behn and Kant, 1999). The underlying foundation of this trend was that competition would result in improved outcomes such as : greater efficiency, higher quality of service, a clearer focus on customers and better value for money (Lavery, 1999; Warner and Hedbon, 2001). In such a direction, the ongoing reforms started all over the world with the separation between the public authority that owns the assets and the supplier (a public, private, or public–private organisation) to which the service provision is contracted out. This separation between the buyer and the supplier, as well as the use of competitive bidding, requires that the service could be clearly specified in a form that can be used as part of the contract and against which performance can be monitored. Due to the increasing customer pressure, the evaluation of the performance of suppliers of public services is gaining more and more importance in the processes of rationalization of public expense aiming at achieving sufficient improvements in service quality (Bovaird, 2006).

ARTICLE IN PRESS 34

A. Ancarani / Journal of Purchasing & Supply Management 15 (2009) 33–42

The concept of service improvement is inherently political and contestable. The performance of public service providers is judged by multiple constituencies. Each of them uses different criteria to judge the standard of public services and may apply different weights to the same criterion. It follows that there is no fixed and universally applicable set of criteria for evaluating whether improvement occurred. Nevertheless, public services have tangible elements that are likely to be valued by all constituencies, even if the valuations differ between groups or over time (Boyne, 2003). Preliminary criteria of service improvement can be derived from the literature on the conceptualisation and measurement of organisational performances in the public sector (Ammons, 2001; Carter et al., 1992). Among such criteria can be considered quantity and quality of outputs, efficiency, equity, outcomes, value for money and customer satisfaction. It is evident that some of the criteria overlap and that customer satisfaction may be considered a proxy of some or all of the above. Actually, from the literature in the leading public administration journals, Boyne (2003) pointed out that the performance measures focus largely on the quantity and quality of outputs and the level of outcomes. Much less is known about influences on efficiency, value for money and customer satisfaction. Due to the lack of studies on the effects of customer satisfaction, it is relevant to investigate how the customers of the public services integrate their service perceptions to obtain a comprehensive valuation of the service quality. In particular, it is necessary to investigate how the expectations, the perceptions of the service and the experiences of lack of coincidence between expected quality (Qe) and perceived quality (Qp) (disconfirmation) in the long-term could influence their evaluation of service quality in terms of the value gained (Bolton and Drew, 1991; Woodruff, 1997). Expectations, performance evaluations and disconfirmation do not necessarily have independent, additive effects for every products according to Oliver (1989) customer responses concerning continuously provided services (as in the case of standardised public services) are characterised by passive expectations and, therefore, disconfirmation will not operate unless the performance is outside the range of experience-based norms. Thus, customers’ assessment of public services continuously provided may depend on performance evaluations only. Even though in the literature there are many methodological proposals for supplier’s performance assessment (among the others for a synthesis De Boer et al., 1998, 2001; Degraeve et al., 2000), however there is a lack of attention toward the public services, even in terms of proper problem formalization. In the present paper, a methodological framework for the assessment of the performance of a public service supplier is provided. In this framework the central role is played by those measures of level of service representative of the Qp by the customers. The applicability of the proposed methodology has been tested on a simplified water service system. An agent-based model has been developed for simulating the interactions amongst the main stakeholders of the system. In particular, in Section 2 the quality of public service has been defined explaining differences and similarities with customer satisfaction, by introducing the concept of value for customer (VC). In Section 3 the proposed methodology for evaluating the supplier of a standardised public services has been presented. In order to apply the methodology, in Section 4, a model of customer value has been developed and, in Section 5, a model of customer satisfaction has been defined in connection with the model of VC. In Sections 6 and 7 the performance indicators were calculated, and the agent-based models for simulating the interactions among the service stakeholders have been synthesised. In Sections 8 and 9 the theoretical case-study has been described

along with some results of the simulations. In Section 10 some final considerations are reported.

2. Quality in public service provision According to the stakeholder salience approach (Mitchell et al., 1997; Neville et al., 2004), there is a positive relationship between number of stakeholder attributes (power, legitimacy, urgency) and perceived stakeholder salience (Parent and Deephouse, 2007). In particular, Kamann (2007) pointed out that public bodies have several potential key-players who may turn, from either dominant or discretionary stakeholders, to definitive stakeholders when one or more stakeholders get attributes (power, urgency) not owned before. It happens in the public sector either when the atomistic stakeholders organize themselves to actually influence the purchasing strategies by public bodies, or, as is the case in many standardised public services, when there is a major change in social feeling that definitively asks for quality. Actually, in the last few years the power and the urgency of the direct customers of the public services are increasing with reference to their capability to claim for their rights. However, just a few changes can be observed in the purchasing strategy by public bodies and even less in public service quality evaluation tools applied. In such a direction, it becomes important to evaluate the role the customer may play both in the policy-development process and in evaluating the policy itself (Hastak et al., 2001). In the public service sector, it is important to consider that the roles of the buyer (e.g. Local Authority), client (customer/citizen) and the supplier (external organisation, public or private) need to be clearly differentiated. The three main actors interact in formal and informal ways, as reported in Fig. 1 where the main relationships are outlined. The triangulation interactions have been widely recognized in the last decades, and hence definitely it is not a new concept. However, what it is quite new is the approach proposed here to evaluate the Qs. While until a few years ago, it was requested to the supplier to provide value for money, nowadays there is an increasing request for VC. Such an approach would require an interpretation of the service quality as a multidimensional concept (Zeithaml, 1988) that should be evaluated in terms of customer’s needs. Traditionally, the customer satisfaction approach has been proposed to evaluate the Qs. However, with particular reference to standardised public service provision, which lasts for a long term, quality cannot be constrained within the customer satisfaction

Provided service Tariffs SUPPLIER

Bills, communications, failures recovery

CUSTOMER / CITIZEN

Contract, complaints, failures communication Promised service

Regulation, contract

Performance indicators

LOCAL AUTHORITY / ATO AUTHORITY

Customer satisfaction

Fig. 1. Relationship among main stakeholders in public services provision.

ARTICLE IN PRESS A. Ancarani / Journal of Purchasing & Supply Management 15 (2009) 33–42

concept (customer’s expectations) since it is strictly referred to a specific transaction (Gro¨nroos, 1984). In order to apply the VC concept, the identification by Local Authorities of the characteristics of service is required that could match the customers/citizens needs. In this respect, the key points are how to identify such needs and how to translate them into technical specifications to be included in contracts with the service providers (supplier). It seems clear that this identification is complex due to the several aspects of the service that need to be considered. Hence, there could be a gap between the performance identified by Local Authority and the performance expected by customers/citizens. The gap could be investigated through a customer satisfaction analysis with respect to specific transactions. However, the customer satisfaction concept cannot match directly with service quality due to the fact that, when a particular perspective is applied, it changes the perception of quality itself. In order to have a comprehensive assessment of the service quality level, it may be necessary to investigate the problem in terms of a broader concept, namely VC (Bolton, 1998; Bolton and Drew, 1991) applied to the public service provision improvement (Boyne, 2003). In this perspective, the customer of the public service becomes the pivot around which the supplier has to build new strategies of development, based on the principles of efficiency, effectiveness and economy of management aiming at attaining adequate performance and at satisfying customer’s requirements. It has to be investigated whether or not such requirements match the level of service agreed between the buyer (Local Authority) and the supplier. Thus, the newness of the approach is the attempt to translate the customer’s requirements and needs into requirements for the supplier.

3. Literature background on customer satisfaction models The foundation of service quality theory lies in the product quality and corresponding customer satisfaction literature. Early conceptualizations (e.g., Gro¨nroos, 1982, 1984; Parasuraman et al., 1985) are based on the disconfirmation paradigm employed in the physical goods literature (Churchill and Surprenant, 1982; Oliver, 1977; Olson and Dover, 1976). The basic concept is that quality results from a comparison between Qp and Qe performance. When such a disconfirmation paradigm has to be adapted to the measurement of service quality, Gro¨nroos (1982) identifies two service quality dimensions, namely functional quality and technical quality. Functional quality represents how the service is delivered; that is, it defines the customer’s perceptions of the interactions that take place during service delivery. Technical quality reflects the outcome of the service act, or what the customer receives in the service encounter. The disconfirmation paradigm also is the basis for Parasuraman et al.’s (1985) SERVQUAL model, which views service quality as the gap between the expected level of service and the customer perceptions of the level received. Although, Gro¨nroos (1982) suggests two dimensions, Parasuraman et al. (1988) propose five: reliability, responsiveness, assurances, empathy and tangibility characteristics of the service experience. In both models the service quality depends on the capability to minimize the gap between Qe and Qp by customers:

DQ ¼ Q e  Q p

(1)

More recently, three themes are evident on service quality. First, several studies advance modified versions of the SERVQUAL model (e.g. Boulding et al., 1993; Carman, 2000; Cronin and Taylor 1992; DeSarbo et al., 1994), paralleled by revised versions proposed by

35

the same original authors (Parasuraman et al., 1991, 1994; Zeithaml et al., 1996). A second theme involves the heightened interest in the technical and functional quality dimensions identified by Gro¨nroos (1982, 1984). Rust and Oliver (1994) offer a three-component model: the service product (i.e., technical quality), the service delivery (i.e., functional quality) and the service environment. The third theme relates to the structure of the service quality construct. Because of the reports of SERVQUAL’s inconsistent factor structure, Dabholkar et al. (1996) identify and test a hierarchical conceptualization of retail service quality that proposes three levels: (1) customers’ overall perceptions of service quality, (2) primary dimensions and (3) subdimensions. This multilevel model recognizes the many facets and dimensions of service quality perceptions (Carman, 1990). In other words, service quality is viewed as a higher-order factor that is defined by two additional levels of attributes.

4. The concept of VC Fekete (1987) argued that value has been conventionally addressed in terms of either its objective meaning (largely within an economic context) or its subjective meaning (as a largely affective, human-based property). It has been evaluated, typically, in a ‘modern’ domain that effectively separates value from values so as to facilitate analysis and measurement, and deliver perceived certainty and exactness. Smith (1987) contested that there is no objective, only contingent value. So, not only does each of us value the same things differently, we individually value different things, and at different times in different ways. From this perspective, value can only be judged within the context of some implicitly limited set of conditions determined by environmental, social and cultural factors as by utilitarian or economic considerations. Consequently, value is intensely personal and exists at a number of different levels. The object and the subject are inextricably connected, and value can be ascribed only at the point of evaluation, or union between the two. Zeithaml (1988) proposed that Qp can be regarded as a consumer overall assessment of the utility of a service based on the perception of what is received and what is given. From this point of view, Qp is the result of the balance between benefits and sacrifices, mostly in terms of trade-off between quality and price. Some Authors have argued that viewing value just as a trade-off between quality and price is too simplistic (Bolton and Drew, 1991). Holbrook and Hirschman (1982) suggested an experiential perspective including the symbolic, hedonic and aesthetic aspects in the consumption process. Sheth et al. (1991) developed a broader theoretical framework of Qp, which regarded the consumer choice as a function of multiple consumption value dimensions. They proposed five dimensions (social, emotional, functional, epistemic and conditional value) contributing to the expanding value constructs. Shillito and De Marle (1992) considered that value is dichotomous, centred in people and the objects they desire. This suggests that value can perhaps be conceptualised and can be best comprehended through the conjoint appreciation of economic and abstract/philosophical perspectives that, together, recognize the existence of value-oriented properties. These properties are associated with both the object and the subject, and are manifest at the point of interaction between the two. Sacrifice and the market are also key factors. The model identifies four distinct interpretations of value (exchange, intrinsic, use and utilitarian), any or all of which may be recognized and/or expressed individually or collectively by the consumer, and all of which

ARTICLE IN PRESS 36

A. Ancarani / Journal of Purchasing & Supply Management 15 (2009) 33–42

are subject to the influence of both the subject’s value system and environmental contingency. Sweeney and Soutar (2000) proposed four distinct value dimensions (emotional, social, quality/performance and price/ value for money). They showed that multiple value dimensions explain consumer choice better than does single value for money item. Holbrook (2006) proposed a typology of customer value composed of extrinsic value versus intrinsic value, and selforiented value versus other-oriented value. Smith and Colgate (2007) developed a framework aiming at identifying four major types of value (functional/instrumental, experiential/hedonic, symbolic/expressive and cost/sacrifice) that can be created by organisations and five major sources of value (information, products, interactions, purchase or consumption environment and ownership/possession) that are associated with central value-chain processes. The resultant 45 table is applied to describe and document customer value creation strategies and serves as a tool for opportunity recognition and product concept specification. From a detailed analysis of the literature, Woodall (2003) identified five primary forms of value for the customer:

of exchange value with intrinsic value and it is essentially utilitarian in nature; this might be a more-or-less objective perception of a tolerable price band (Liljander and Strandvik, 1992), and/or a market price (Anderson, 1995), and/or a maximum or reservation price (Anderson, 1995). Woodall (2003) identified many specific types of value associated with his higher order: derived VC, marketing VC and sale VC constructs. Some limitations in his study may be outlined, as there is considerable overlap in the categories in the sense that the same benefits appear under multiple headings. Moreover, Woodall (2003) did not identify the sub-dimensions of customer value of which the specific benefits and sacrifices might be illustrative examples. In analysing the quality in public services, where three main actors interact with each other continuously, the final goal is to evaluate how much the value provided by the supplier matches the value required by both the buyer and the customers. Since the buyer’s institutional duty is to satisfy the customer’s needs, two kinds of analysis can be carried out: the first aiming at comparing the value provided by supplier with the one required by the customer and the second aiming at determining the gap between customer expectations and buyer perceptions of those expectations.

 net VC, in terms of balance of benefits and sacrifices: it implies



 



that the customer makes some judgment on the usefulness of a product by computing or comparing benefits and sacrifices; some authors ascribe it to a ratio or to the dividing of benefits by sacrifices (e.g. Heskett et al., 1997); other writers consider the computation to be a matter of subtracting sacrifices from benefits (e.g. Lai, 1995); some authors acknowledge that there is some form of relational comparison that might relate to an individual product alone, or between it and competing alternatives (e.g. Fornell et al., 1996); all recognize that coincident consideration of both benefits and sacrifices is essential for establishing a sense of VC, though there is less consensus regarding whether accrued benefits should be perceived as product attributes (e.g. quality, performance), or as outcomes (use, convenience), or as both; or whether relevant sacrifices are entirely practical/cognitive (e.g. cost), entirely of the senses/affective (disappointment) or a binary amalgam; it has potentially different subforms; derived VC, in terms of use/experience outcomes: it is suggestive of the notion of use value (e.g. Sheth et al., 1991); all are substantially informed by the linking of consumption experiences to social (Kahle, 1983) and human (Rokeach, 1973) values; VC here is conceptualised as the benefits derived from consumption-related experience and is presented such that independence of, or at least prevalence over, any sense of associated sacrifice is implied; marketing VC, in terms of perceived product attributes: this view of VC perhaps favours a supplier-oriented perspective (e.g. Treacy and Wiersema, 1995); sale VC, in terms of reduction in sacrifice or cost: it is oriented to a demand-side value interpretations (Zeithaml, 1988), or as one of a number of potential product attributes (Dodds, 1999); this value means low relative price within a competitive environment (market), and can be viewed, in part, as being analogous to exchange value; it is associated with reduction of sacrifice than it is with increase in monetary gain, and here ‘best’ value is delivered by the lowest-priced alternative; neither use nor the balancing of benefits and sacrifices nor the nature of product attributes impacts substantially upon this particular interpretation of VC; rational VC, in terms of assessment of fairness in the benefit–sacrifice relative comparison: it combines the notions

5. A methodological approach The methodology here proposed is based on the definition of service quality as a compendium of functional and technical quality of service (Gro¨nroos, 1984). It means that the service quality has to be evaluated with reference both to the contract’s specifications, reported in the contract between the buyer and the supplier, and to the Qe and Qp by customers (Fig. 2). Ancarani and Capaldo (2001) defined a general framework for the evaluation of the quality of public services. The methodology here proposed builds on the previous framework and aims specifically at assessing the performance of the supplier of a

1st STEP Analysis of present legal framework

Stakeholders identification

Analysis of service standard

Analysis of information flows

Definition of main service processes

Identification of institutional duties

Identification of main activities

2ndSTEP Model of value for customer

rd

Level of service to be guaranteed by L.A.

Model of customer satisfaction

Identification technicalorganisational indicators

Identification stakeholders’ interaction indicators

3 STEP Simulation of perceived service level in terms of value for customer Fig. 2. Methodological framework for supplier’s evaluation.

ARTICLE IN PRESS A. Ancarani / Journal of Purchasing & Supply Management 15 (2009) 33–42

standardised public service. It can be divided into three main steps. The first step regards the identification of minimum standards of service that must be guaranteed. It would represent the Qs by contract. Such an analysis would require identifying the stakeholders involved and the informational flows among them, as well as the operational processes of the analysed service. In the second step, the methodology requires identifying a model of value for client/customer that includes progressive gains of value along both the experience cycle with the service and the interaction cycle with the service provider. The definition of the model of VC has to be paralleled by the identification of an appropriate set of performance indicators representative of the service level the Local Authority may guarantee. The value attributes identified by the model and the measures provided by the indicators should match. However, in several cases, they could diverge due to the gaps related to information asymmetry, lack of capability in interpreting the customer’s needs and contract’s incompleteness. The identification of the indicators should be done in order to compare both the quality provided with the quality required by contract (Qg), and the Qe with the Qp by the customers. Such indicators have to take into account the technical-organisational aspects as well as the interactions among the stakeholders of the system. These comparisons require the definition of a reference scheme for customer satisfaction that relates simultaneously to the three main stakeholders involved (customer, supplier and Local Authority). If the indicators are identified taking into account the model of VC, the whole set of indicators provides progressive and marginal values that the customer obtains along both the experience cycle with the service and the interaction cycle with the supplier. In the third step of the methodology, a set of behavioural rules for the supplier should be defined, along with a set of behavioural rules for the customer, and the interaction rules between the supplier and the customer. Such last rules represent the existing relationships between the technical-organisational parameters of the service provided and the service levels perceived by the customer. The rules have to describe how the VC impacts on the supplier’s behaviour and how the customer satisfaction evolves along the service provision experience cycle.

37

time phases of interaction, experience cycle with the service and experience cycle with the supplier. Traditionally, the net value has been considered sufficient for evaluating the customer perception of service quality. However, it implies that the contractual relationship would be time constrained with a transaction and a recession. Actually, this barely happens in the standardised public services. Therefore, a more appropriate value model is necessary. In the case of standardized public services an aggregated value model could be appropriate, since it allows appreciating how customer perceptions change during the service experience phases. Moreover, in order to take into account the triangulations among the three main stakeholders, the model of VC has to be different in the two cases when the customer is either the Local Authority, the buyer who entrusts with the service provision the supplier (also the customer who contributes to the service payment with the asset availability) or the citizen who makes use of the service and directly pays for the service consumption. The time duration and the relevance of the phases are different for the two typologies of customers mentioned above. The relevant time phases for a generic public service are reported in Table 1. Combining the forms of value and the phases along the service experience, the model of value can be sketched as a Rubik cube in which each small face represents one of the components of Qp (Fig. 3). In particular, the model is appropriate for the standardized public services as it takes into account both the longitudinal perspective (temporal dimension of value for the customer) and the transversal perspective (value perceptions change during the transition from one experience phase to another). The components of the aggregated value applied in the model are reported in Table 2. It has to be noted that the model cannot be applied directly both for service to people, where the service’s outcome can be appreciated only after a long period, and for network services, where the pre-purchase, contract and disposition phases are extremely relevant.

Net VC

Rational VC

Marketing VC Marketing VC

Sale VC

Emergency

Long - term operations

6. The concept of value for the customer

Derived VC

6.1. A model of VC In order to measure the level of service perceived by customers, it is necessary to formalize an appropriate model of VC. In the present paper Woodall model (2003) acts as a reference model in terms of measurement of service value as an aggregated amount including all the customer perceptions along the different

Recovery PREPURCHASE

CONTRACT

BILLING

ROUTINE OPERATIONS

FAILURE/ RECOVERY RE-AGREEMENT/ DISPOSITION

Fig. 3. Model of aggregated VC for standardised public services.

Table 1 Time and experience phases along the public service provision. Time phases

Experience phases

Local authority

Citizen

(1) Ex ante (pre-purchase) (2) Transaction (at the point of contract and during experience) (3) Routine operations (normal operation) (3 bis) Failure/recovery from failures (during use/experience) (4) Re-agreement or disposition

Collecting information Purchase Learning Emergency Purchase/disposition

Extremely relevant Relevant/ N.A.b Normal Normal Relevant

Not relevanta Normal/relevant Relevant Extremely relevant N.A.c

a b c

Very often only one supplier is available. The L.A. do not pay the tariff. It rarely happens for citizens.

ARTICLE IN PRESS 38

A. Ancarani / Journal of Purchasing & Supply Management 15 (2009) 33–42

phases. The application of the model separately to the two customers (citizen and Local Authority) of a standardised public service makes clear how different is the impact of the phases on the customer value in the two cases (Figs. 4 and 5). Moreover, such differences point out that the two final values can differ significantly, thus determining a gap in both expectations and perceptions between citizen and Local Authority. The two applications differ mainly because of the major contribution to aggregated value gained during the routine operations by the customer/citizen, while substantial contributions are present also in the other phases when the customer is the Local Authority. Since the present analysis is focused on a model of value for citizen/customer, it is useful to analyse in some more detail the value he can gain in each phase, as represented in Fig. 4. In phase 1, the VC can be expressed in terms of the existence of information about the quality of the provision and the capability by front-end people to quickly reply to customers about time connection or payment required for new connection. In phase 2, the value depends on the amount of sacrifices required in terms of organisational time (front-end office interactions) and ‘‘technical’’ time (time for supply activation). Finally, in phase 3, the value depends on the quality of water supplied, the precision of meter reading (efficiency in billing), the time for interacting with the supplier (waiting time on the phone, on the window), time between complaints forwarded and replied by the supplier, technical and organisational capabilities of the supplier to guarantee continuity of service, to notice users when an interruption occurs (planned maintenance), to play a proper role when a failure occurs (unplanned maintenance), that can bring any kind of risk for customer wellness in general. As already noted, the fourth phase does not add any value due to the lack of the disposition option.

In the model, the rational value includes the costs for the customers to collect information about the available options. Since the local public services are often provided under monopolistic conditions, such a value can be associated only with the information search focused to complaint for an unexpected tariff increment. In fact, it is generally addressed to the Local Authority and substitutes the disposition from the service. The model of VC can be applied by considering aggregated value cube projections on planes corresponding to the time

Table 2 Value forms. Value forms

Value form components

Type of net value acquired or paid for

Marketing

Technical quality Core features Organisational, economical, relational quality Features of added services/ customization

Benefit—attribute Benefit—attribute Benefit—attribute

Rational

Opportunity costs

Sacrifice—monetary

Sale

Price Acquisition costs Opportunity costs Maintenance costs Relationship/psychological costs Time

Sacrifice—monetary Sacrifice—monetary Sacrifice—monetary Sacrifice—monetary Sacrifice—non monetary Sacrifice—non monetary

Derived

Personal benefits Social benefits Practical benefits

Benefit—outcome Benefit—outcome Benefit—outcome

PREPURCHASE

Benefit—attribute

CONTRACT

Marketing VC

ROUTINE OPERATIONS

Marketing VC

Sale VC

FAILURE/RECOVERY Marketing Net VC VC

Sale VC

Sale VC

Derived VC

RE- AGREEMENT/ DISPOSITION

Derived VC FAILURE/RECOVERY Net VC

Marketing VC

Sale VC

Derived VC

Fig. 4. Application of the model of value for citizen.

PREPURCHASE Marketing VC

CONTRACT Marketing VC

Marketing VC

Sale VC

FAILURE/RECOVERY Rational VC

Sale VC

RE-AGREEMENT/ DISPOSITION

ROUTINE OPERATIONS Derived VC

Net VC

FAILURE/RECOVERY

Net VC

Marketing VC

Net VC

Marketing VC

Sale VC

Derived VC

Sale VC

Derived VC

Marketing VC Rational VC Sale VC

Fig. 5. Application of the model of value for Local Authority.

ARTICLE IN PRESS A. Ancarani / Journal of Purchasing & Supply Management 15 (2009) 33–42

In calculating Qs and Qg the following has been hypothesised (Gro¨nroos, 1984):

6.2. Quality model adapted to standardised public services In the present paper, the PZB model (Parasuraman et al., 1985) has been adapted to the local public service sector by taking into account the three main actors involved in the service provision (Fig. 6). The quality gap the Local Authority should monitor with the supplier can be defined as follows:

DQ C ¼ Q g  Q s

(2)

The quality gap the client/customer requires to overcome can be defined as follows:

DQ CUS ¼ Q e  Q p

(3)

There are several reasons why DQC may differ from DQCUS. The Qg can differ from Qe due to the different perspectives and goals by Local Authority and the customer:

DQ GE ¼ Q g  Q e

(4)

Moreover, Qg depends on the supplier’s market power (in terms of imagine, financial availability, expertise) and by informative asymmetry between the buyer and the supplier. Theoretically speaking, in local public services such a difference could be considered unimportant (Hp. 1). However, in practice the environmental specific conditions might not allow such an evaluation. Furthermore, either the supplier’s opportunistic behaviour, with reference to the verification of contractual agreement, or the lack of monitoring capability by the buyer could determine a high gap between the two values. The Qs may differ from Qp due to the lack of effective communication (DQCOM) between the supplier and the client/ customer, as already stated by the PZB model:

DQ COM ¼ Q s  Q p

(5)

In local public services, such an informative gap could be influenced by the relationships between the buyer and the supplier. Some investigations carried out on water service and on road service have pointed out the small gap between customer perceptions and buyer goals. Hence, it could be considered reasonable to skip this gap (Hp. 2). Wherever the two hypothesis are contemporaneously applied, the above-mentioned triangulation among the buyer, the supplier and the client/customer would resolve in a dyadic relationship buyer–supplier. Such an assumption can be considered reasonable for some services, but needs to be analysed case by case, while it is not applicable to other services:

DQ C ¼ Q CUS

39

(6)

In our case, for the sake of generality, the model maintains separately the two Equations. (2) and (3).

 Qs is the total quality level measured as a weighted mean



between technical and functional qualities; technical quality can be measured by means of technical performance indicators while functional quality can be measured by means of organisational performance indicators; Qg is defined according to the legislative standards, particularly with reference to the minimum level defined in the Service Chart.

The difference in Eq. (6) represents the total gap between Qp by buyers and Qp the by customers, when DQCOM ¼ 0 and DQGE ¼ 0. The difference in Eq. (2) can be calculated by means of the performance indicators monitored according to the contract. The difference (3) can be calculated by means of a survey among the customers, for example according to the SERVQUAL approach (Parasuraman et al., 1991). The gap between Qe and Qs can be calculated as follows: Q e  Q s ¼ DQ C  DQ GE  DQ COM

(7)

The gap between Qg and Qp can be calculated as follows: Q g  Q p ¼ DQ CUS  DQ GE  DQ COM

(8)

All the identified gaps among the stakeholders of the system are relevant for improving the management and provision of local public services. 7. Performance indicators The performance indicators represent a useful tool for evaluating the service performances. The main stakeholders involved in the provision of standardised public services are as follows:

 Supplier: a public, private or public–private partnership organisation who is entrusted with service provision;

 Customers: the direct users of the service that have a customer–provider relationship with water service supplier;

 Local Authority/ies: focused on supplier’s evaluation, guaran 

teeing the respect of the contractual agreements also with regard to customer expectations; Public organisms: entrusted with regulation, control and supervision, also responsible for supplier’s monitoring and evaluation; Indirect users: not directly involved in the service but affected by the supply process with reference to environmental consequences; in particular, they may include public opinion and proactive users (environmental organisation, pressure groups, etc.).

The above-mentioned stakeholders interact in different environments as follows:

BUYER

 Politic-institutional environment: it is responsible for sector Hp1

QC Qs

policy at local, regional and national levels;

Qe QCUS

Hp2 QCOM

Qp

CUSTOMER

SUPPLIER

Qg

QG-E

Fig. 6. Representation of the perspectives on service quality by the buyer, the supplier and the customer.

 Socio-economic environment: it provides funds for investments;  Natural environment: it includes the physical environment and might include the natural resources extracted. The set of performance indicators has to be identified in each case with respect to the technical-organisational aspects and to the interactions between either the customer and the supplier or the buyer and the supplier, as requested by the methodology proposed in this paper.

ARTICLE IN PRESS 40

A. Ancarani / Journal of Purchasing & Supply Management 15 (2009) 33–42

The identification of an appropriate set of indicators would require taking into account first the guidelines proposed by institutional organisms. As usual, the indicators must reflect what is required by law mainly with reference to the minimum standard of service to be guaranteed to the customers. Other important documents that have to be taken into account are the formal agreements between Local Authority and the supplier and the contract between the customer and the supplier.

Table 3 Cognitive nature and behaviour of the agents. Classes

Cognitive nature

Objectives

Algorithm of action

Customer

Hybrid

Complaint if something ‘‘does not work’’

1. Control of state variable 2. Classification of contingency (gradually) 3. Update of internal state function 4. Complaint or not

Supplier

Hybrid

Reply to complaints ‘‘technically and organizationally’’ to minimize it and maximizing customer satisfaction

1. Control of budget internal state function 2. Control of user’s complaints 3. Classification of contingency 4. Resolving conflict situations (corrective actions according to the priority rate k) 5. Implementation of corrective actions through budget consumption 6. Update of budget internal state function

8. Evaluation of the service by means of an agent-based simulation The agent-based models add to calculability, typical of mathematical models, and flexibility, typical of descriptive models. Moreover, they allow running simultaneously analysis both on parts and on the whole system simulated (e.g. Holldobler and Wilson, 1990). Wooldridge and Jennings (1995) classified agents into three groups, according to cognitive nature: reactive, deliberative and hybrid. Reactive agents can only have pre-defined behaviours to certain inputs, without taking into account where the whole system is going. Deliberative agents are thinking agents that elaborate complex plans of action combining many simple behavioural schemes. They take into account their internal state and try to predict effects of their actions before they act. To evaluate the supplier’s performance a multi-agent based (MAS) can be built. A MAS acts as a dynamic multi-criteria system for decision making. By coding each decisional scenario, choosing an adequate cognitive nature for agents, modelling environment and selecting proper coordination mechanisms, it is possible to unify different perspectives and try to reach an optimal solution for the whole system and not only for one or some parts. Such advantages seem to fit the goal of the present paper since the agent model allows considering both the different aspects of service evaluation (functional and technical service quality) and the different perspectives of customers and Local Authorities. Moreover, it allows the non-conflicting co-existence and mixing of different value models in an aggregated measure (aggregated value). The model is composed by a structured environment—where the autonomous agents interact—able to transfer information on performances (from supplier to customers) and complaints (from customer to supplier) with the possibility of considering a neutral observer who is monitoring these interactions (Local Authority). The system can be simulated through two classes of autonomous agents, namely supplier and customer. The class of supplier is always made of a single element. The class of users represents the citizens served by the supplier and, depending on the local conditions, it may be divided into different subclasses. Each class of agents is characterized also by its cognitive nature, which influences the behaviour, and by the state variables. Since the present paper deals with standardised public services, it makes sense that an operational excellence strategy would be required. However, the characteristics and criteria for supplier evaluation in this case may be completely different from the ones required by services to the people characterized by customer intimacy strategy. Different strategies adopted by the Local Authority would lead to the definition of different typologies of agents with characteristics adequate to the investigated public service. Furthermore, the cognitive nature and the behaviour of the agents will be different according to the market conditions in which the public service is provided. Monopolistic markets will ask for more deliberative approaches from both customer and supplier agents, while more reactive behaviour is expected from supplier agents in competitive markets.

Customer 1

Provided service

Provided service

Supplier Complaints

Customer N

Complaints

Fig. 7. Simulation scheme and Informative flows (interactions scheme).

In Table 3 the cognitive natures of the agents in the case of standardised public services, provided in a competitive market (at least with a competition for the market even if not in the market), are described. The customer agent is hybrid because it reacts to any misfit occuring during service provision (reactive). However, it gradually recognizes any kind of contingency and fits his level of satisfaction to real conditions of supply (deliberative). The supplier agent is hybrid because it knows the corrective actions necessary to improve the level of service (reactive) but elaborates plans of action by classifying contingencies, through a priority rate (deliberative). The simulation scheme and the informative flow among the agents could be modelled as reported in Fig. 7. Each cycle of the simulation (j ¼ 1, y, 365) starts with monitoring the customer value Qp by the users; it depends on the value attained by the state variables. Each customer (it can represent a single user or a group of users experiencing the same service level) compares its internal state, component by component (CSi), with the standard value fixed by regulation (CSobji ¼ minimal level of service for each component). According to the definition of minimum level of service to be guaranteed in order to avoid dissatisfaction (see Section 7), from this comparison the customer computes a gap, representative of its lack of satisfaction for each variable, as follows:   CSi  CSobji DðCSi Þj  CSobji j

ARTICLE IN PRESS A. Ancarani / Journal of Purchasing & Supply Management 15 (2009) 33–42

41

Table 4 Changes of independent variables. Simulated independent variable (PI2)

Optimal value

Range to stay optimal

Percentage loss

Updated value

TOk

TOk

TO01 ; TO11 

Dk

ðTOk Þj ¼ ðTOk Þj ð1  Dk Þ

All the computed gaps are then compared to the fixed bound values according to the characteristics of the service investigated. At the end of each cycle ‘‘j’’, users update their global internal satisfaction state taking into account the value of whole D(CSi)j. The global customer satisfaction at cycle ‘‘j’’ is at maximum level only if D(CSi)jXBound1 8i. The supplier, when receiving a complaint, controls its financial wellness in terms of budget. If its overall value is greater than zero, it decides to act or not depending on priority rate ‘‘k’’. The priority rate k is computed by the Rule Master, a simple neural network, as follows: 1. 2. 3. 4.

at time ‘‘j’’ any type of misfit is additively computed; trial performance indicators are built on these measurements; trial PI are compared to complaints register of the supplier; each state variable has high priority if both correlated failures and complaints are high; 5. the supplier controls costs and times of corrective actions, from higher to lower priority and decides to act optimizing its budget function; 6. the Rule maker updates the state variables in terms of priority rates (connections weights). At the beginning of each simulation the components of customer satisfaction (CSi)j are equal to CSobj fixed by contract for each customer. For each technical-organisational variables (TOk), according to the classical model of failures following a Poisson distribution, the probability of failure is computed through a negative exponential distribution formula: f(t) ¼ elt whose parameters are calculated with respect to the historical data available. For each TOk a random number is created and a failure occurs if rndA[f(t);1], while no failure occurs in [0; f(t)[. The simulated value CSi is optimal (CSobji) if the related PI2 variables are in their optimal ranges. Otherwise CSi ¼ (CSi)j1[1D(CSi)] with D(CSi) equal to the percentage loss on CSi. In Table 4 it is reported how the independent variables change with respect to the fixed standard. The actual response of the system can be analysed once the indicators and the relationships among them and the customer satisfaction are defined for each specific service investigated.

9. Conclusions The evaluation of the performances in the public sector is increasingly important, due to the trend to contracting out the provision of several local services to organisations external to the Local Authorities that own the assets. In such services the evaluation must take into account both the technical parameters (resulting from the interaction between the buyer and the supplier) and the customer satisfaction (resulting from the interactions between the supplier and the customer). In the present paper, a methodology for evaluating the performance of the supplier of a standardised public service is presented with a theoretical application to a generic standardised public service. It aims at supporting the managers of the Local Authorities in their efforts to improve the service provided including in the service requirements the customer satisfaction.

The basic concept is that. the quality of the service depends on the quality of the supplier’s performances. It is then possible to determine such performances in terms of technical and functional quality as linked with customer satisfaction. If customer satisfaction can be expressed in terms of a prevalent value model for the users, the quality of interaction between the customer and the supplier is a function of the quality of the service provided. The agent-based model developed for simulating the interactions between the customer and the supplier shows an interesting bivalence. It allows the Local Authority to select from different potential suppliers the one having the best investment plan to maximize customer satisfaction (forecast simulation), and it also allows monitoring the supplier and assess its performances (customer satisfaction) with reference to contingencies and corrective actions implemented. Several limitations must be recognized to the reported example. The results depend on the scheme implemented in terms of both technical characteristics and interaction rules applied. In particular, the rules are very stringent as both the supplier aims at satisfying the complaining and the customer complaints immediately trusting in the supplier’s ability to improve the system. It is not considered the case, indeed frequent, that the customer does not complain at all. The PI reference values have been defined according to the minimum level of service. The analysis of a real case could test the model applicability. Moreover, the customer satisfaction simulated through the model should be compared with the result of a field survey aiming at verifying whether it actually matches the customer perceptions. References Ammons, D., 2001. Municipal Benchmarks. Sage, London. Ancarani, A., Capaldo, G., 2001. Management of standardised public services: a comprehensive approach to quality assessment. Managing Service Quality 11 (5), 331–341. Anderson, E.W., 1995. An economic approach to understanding how customer satisfaction affects buyer perceptions of value. Proceedings of the AMA Winter Educators Conference 6, 102–107. Behn, R.D., Kant, P.A., 1999. Strategies for avoiding the pitfalls or performance contracting. Public Productivity & Management Review 22 (4), 470–489. Bolton, R., 1998. A dynamic model of the duration of the customer’s relationship with a continuous service provider: the role of satisfaction. Marketing Science 17 (1), 45–65. Bolton, R., Drew, J., 1991. A multistage model of customer assessment of service quality and value. The Journal of Consumer Research 17 (4), 375–384. Boulding, W., Kalra, A., Staelin, R., Zeithaml, V.A., 1993. A dynamic model of service quality: from expectations to behavioural intentions. Journal of Marketing Research 30 (February), 7–27. Bovaird, T., 2006. Developing new forms of partnership with the ‘market’ in the procurement of public services. Public Administration 84 (1), 81–102. Boyne, G.A., 1998. Competitive tendering in local government: a review of theory and evidence. Public Administration 76 (4), 695–712. Boyne, G.A., 2003. Sources of public service improvement: a critical review and research agenda. Journal of Public Administration Research and Theory 13 (3), 367–394. Brown, T.L., Potoski, M., 2003. Contract-management capacity in municipal and county governments. Public Administration Review 63 (2), 153–164. Carman, J.M., 1990. Consumer perceptions of service quality: an assessment of the SERVQUAL dimensions. Journal of Retailing 66 (1), 33–55. Carman, J.M., 2000. Patient perceptions of service quality: combining the dimensions. Journal of Services Marketing 14 (4), 337–352. Carter, N., Day, P., Klein, R., 1992. How Organizations Measure Success. Routledge, London.

ARTICLE IN PRESS 42

A. Ancarani / Journal of Purchasing & Supply Management 15 (2009) 33–42

Churchill, G.A., Surprenant, C., 1982. An investigation into the determinants of customer satisfaction. Journal of Marketing Research 19 (November), 491–504. Cronin, J.J., Taylor, S.A., 1992. Measuring service quality: a reexamination and extension. Journal of Marketing 56 (July), 55–68. Dabholkar, P., Thorpe, D.I., Rentz, J.O., 1996. A measure of service quality for retail stores. Journal of the Academy of Marketing Science 24 (Winter), 3–16. De Boer, L., van der Wegen, L., Telgen, J., 1998. Outranking methods in support of supplier selection. European Journal of Purchasing and Supply Management 4 (2), 109–118. De Boer, L., Labro, E., Morlacchi, P., 2001. A review of methods supporting supplier selection. European Journal of Purchasing and Supply Management 7 (2), 75–89. Degraeve, Z., Labro, E., Roodhooft, F., 2000. An evaluation of vendor selection models from a total cost of ownership perspective. European Journal of Operational Research 125 (1), 34–58. deLeon, L., Denhart, R.A., 2000. The political theory of reinvention. Public Administration Review 60 (2), 89–97. DeSarbo, W.S., Huff, L., Rolandeli, M.M., Choi, J., 1994. On the measurement of perceived service quality: a conjoint analysis approach. In: Rust, R.T., Oliver, R.L. (Eds.), Service Quality: New Directions in Theory and Practice. Sage Publications, Thousand Oaks, CA, pp. 201–222. Dodds, W.B., 1999. Managing customer value. Mid-American Journal of Business 14 (1), 13–22. Erridge, A., Greer, J., 2002. Partnerships and public procurement: building social capital through supply relations. Public Administration 80 (3), 503–522. Fekete, J. (Ed.), 1987. Life after Postmodernism: Essays on Value and Culture. New World Perspectives, Montreal. Filkin, G., 1998. Starting to Modernize: The Change Agenda for Local Government. The Joseph Rountree Foundation, New York. Fornell, C., Johnson, M.D., Anderson, E.W., Jaesung, C., Bryant, B.E., 1996. The American customer satisfaction index: nature, purpose and findings. Journal of Marketing 60 (October), 7–18. Gro¨nroos, C., 1982. An applied service marketing theory. European Journal of Marketing 16 (7), 30–41. Gro¨nroos, C., 1984. A service quality model and its marketing implications. European Journal of Marketing 18 (4), 36–44. Hastak, M., Mazis, M.B., Morris, L.A., 2001. The role of consumer surveys in public policy decision making. Journal of Public Policy & Marketing 20 (2), 170–185. Heskett, J.L., Sasser Jr., W.E., Schlesinger, L.A., 1997. The Service Profit Chain. The Free Press, New York. Holbrook, M.B., 2006. Consumption experience, customer value, subjective personal introspection: an illustrative photographic essay. Journal of Business Research 59 (6), 714–728. Holbrook, M.B., Hirschman, E., 1982. The experiential aspects of consumption: consumer fantasies, feelings, and fun. Journal of Consumer Research 9 (2), 132–141. Holldobler, B., Wilson, E.O., 1990. The Ants. Harvard University Press, Cambridge, MA. Kahle, L.R. (Ed.), 1983. Social Values and Social Change. Praeger, New York. Kamann, D.J.F., 2007. Organizational design in public procurement: a stakeholder approach. Journal of Purchasing and Supply Management 13 (2), 127–136. Kettl, D.F., 1993. Sharing Power. The Brookings Institute, Washington DC, USA. Lai, A.W., 1995. Consumer values, product benefits and customer value: a consumption behaviour approach. In: Kardes, F.R., Sijan, M. (Eds.), Advances in Consumer Research. Association for Consumer Research, pp. 381–383. Lavery, K., 1999. Smart Contracting for Local Government Services: Processes and Experience. Praeger, Westport, CT. Liljander, V., Strandvik, T., 1992. Different types of customer service expectations in relation to perceived value. Proceedings of the Annual Conference—European Marketing Academy 21 (2), 1351–1354. Milward, H.B., 1996. Symposium on the hollow state: capacity, control and performance in interorganizational settings. Journal of Public Administration Research and Theory 6 (2), 193–195.

Milward, H.B., Provan, K.G., Else, B.A., 1993. What does the hollow state look like? In: Bozeman, B. (Ed.), Public Management: the State of the Art. Jossey-Bass, San Francisco, CA. Mitchell, R.K., Pyle B, B.R., Wood, D.J., 1997. Towards a theory of stakeholder identification and salience: defining who and what really counts. Academy of Management Review 22 (4), 853–886. Neville, B.A., Bell, S.J., Whitwell, G., 2004. Stakeholder salience revisited: toward an action tool for the management of stakeholders. In: Academy of Management Best Conference Paper, SIM D1–D5, 2004. Oliver, R.L., 1977. Effect of expectation and disconfirmation on post-expense product evaluations: an alternative interpretation. Journal of Applied Psychology 62 (4), 480–486. Oliver, R.L., 1989. Processing of the satisfaction response in consumption: a suggested framework and research propositions. Journal of Consumer Satisfaction, Dissatisfaction and Complaining Behaviour 2, 1–16. Olson, J.C., Dover, P., 1976. Effects of expectation creation and disconfirmation on Belief elements of cognitive structure. In: Anderson, B.B. (Ed.), Advances in Consumer Research, vol. 3. Association for Consumer Research, Chicago, pp. 168–175. Parasuraman, A., Zeithaml, V.A., Berry, L.L., 1985. A conceptual model of service quality and its implications for future research. Journal of Marketing 49 (4), 41–50. Parasuraman, A., Zeithaml, V.A., Berry, L.L., 1988. SERVQUAL: a multiple item scale for measuring consumer perceptions of service quality. Journal of Retailing 64 (1), 12–37. Parasuraman, A., Zeithaml, V.A., Berry, L.L., 1991. Refinement and reassessment of the SERVQUAL scale. Journal of Retailing 67 (4), 420–450. Parasuraman, A., Zeithaml, V.A., Berry, L.L., 1994. Reassessment of expectations as a comparison standard in measuring service quality: implications for future research. Journal of Marketing 58 (February), 201–230. Parent, M., Deephouse, D., 2007. A case study of stakeholder identification and prioritization by managers. Journal of Business Ethics 75 (1), 1–23. Rokeach, M., 1973. The Nature of Human Value. The Free Press, New York. Rust, R.T., Oliver, R.L., 1994. Service quality: insights and managerial implications from the frontier. In: Rust, R.T., Oliver, R.L. (Eds.), Service Quality: New Directions in Theory and Practice. Sage Publications, London, pp. 1–20. Sheth, J.N., Newman, B.I., Gross, B.L., 1991. Why we buy what we buy: a theory of consumption values. Journal of Business Research 22, 159–170. Shillito, M.L., De Marle, D.J., 1992. Value: Its Measurement, Design and Management. Wiley & Sons, New York. Smith, B.H., 1987. Value without truth-value. In: Fekete, J. (Ed.), Life after Postmodernism: Essays on Value and Culture. New World Perspectives, Montreal. Smith, J.B., Colgate, M., 2007. Customer value creation: a practical framework. Journal of Marketing Theory and Practice 15 (1), 7–23. Sweeney, J.C., Soutar, G.N., 2000. Consumer perceived value: the development of a multiple item scale. Journal of Retailing 77 (3), 203–220. Treacy, M., Wiersema, F., 1995. The Discipline Of Market Leaders: Choose Your Customers, Narrow Your Focus, Dominate Your Market. Harper Collins, London. Warner, M., Hedbon, R., 2001. Local government restructuring: provatization and its alternatives. Journal of Policy Analysis and Management 20 (2), 315–336. Woodall, T., 2003. Conceptualising v value for the customer: an attributional, structural and dispositional analysis. Academy of Marketing Science 12, 1–42. Woodruff, R.B., 1997. Customer value: the next source of competitive advantage. Journal of the Academy of Marketing Science 25 (2), 139–153. Wooldridge, M., Jennings, N.R., 1995. Intelligent agents: theory and practice. Knowledge Engineering Review 10 (2), 115–152. Zeithaml, V.A., 1988. Consumer perceptions of price, quality and value: a meansend model and synthesis of evidence. Journal of Marketing 52, 2–22. Zeithaml, V.A., Berry, L.L., Parasuraman, A., 1996. The behavioral consequences of service quality. Journal of Marketing 60 (April), 31–46.