Business process outsourcing enhanced by fuzzy linguistic consensus model
Graphical abstract
Introduction
In this turbulent era, characterized by high complexity, business organizations, to remain competitive, should enlarge their boundaries, creating relationships with others; value creation, in fact, is no longer realized by the only producer, but it is co-created by all the subject involved in a service network [1]. In this perspective, there is an urgent need for companies to focus on their core competences, entrusting entire business processes to specialized service providers. In fact, firms should invest in the planning and design of new products or services and in the organization of extended supply chain in an effort to improve the creation of value for consumers, remaining competitive at the same time. Value Chain represents a set of operations that a firm executes to deliver products or services to the market [2]. A single company usually relies on more than one value chain. The concept of the value chain is now placed side by side with the concept of Value Net that is introduced to emphasize the inability of a company (and thus for a single value chain) to compete on the market alone [3]. Largest companies own several large-scale processes requiring efficiency and tight integration, others companies have processes that change rapidly generating highly customized outputs. In the first case, it is important to formally model processes to better address optimization in terms of high quality and low cost. In the second case, according to a strategy emphasizing flexibility, it could not be the right choice to model rigid value chains. Unlike value chain, a value net typically cannot produce highly efficient outputs [3] but it can better adapt itself to highly dynamic strategies, goals, and customers. Consequently, business organizations open toward the outside to include other value chains, creating alliances with other firms and delegating their processes to specialized companies. Therefore, these networks of relationships entail the need for the establishment of a collaborative management, leading companies to outsource their business processes as a tool for obtaining the overall improvement of performances. Business process outsourcing (BPO) is the set of practices contemplating the entrusting of business processes to other organizations. BPO constitutes a strategic choice that companies can implement in order to establish stable partnerships with other companies based on the sharing and co-definition of the objectives. The strategic nature of this process, in fact, has been translated into the common practice of externalizing core business processes, such as production, logistics, marketing, design, R&D and human resources.
The adoption of BPO brings out human decision-making problems. In fact, the effectiveness of this strategic option opens the way for complex decision-making processes, especially during the critical moment of service provider selection and choose. In this context, different business process owners should define not only the most fitting parameters to gain a full compliance between outsourcing and in-house processes, but above all, they should harmonically converge towards the criteria at the roots of decisions. It is evident that this process is very complex, because different skills, expertise, experiences, and knowledge characterize business decision-makers. Managerial literature is still calling for a decision-making process [4], also underlining the importance of qualitative issues (e.g., trust, values, ethics) useful to offer those conditions of full strategic and organizational adherence of value chain that have to be involved in BPO. The governance plays a key role in establishing models, methods, rules, and tools for the resolution of such problems. Thus, decision-makers have to consider the alignment between organizational priorities and business processes fostering constant and effective action to enhance performance [5]. Indeed, business processes should be designed and evaluated considering the strategic priorities, creating a bidirectional relationship between corporate strategies (guidelines and objectives) and operational processes. According to the aforementioned considerations, this work proposes the definition of a framework that mainly focuses on human decision-making activities within value nets and, in particular, it proposes a formal model to support group decision-making, taking care of both the context in which decisions have to be made and the heterogeneity (in terms of knowledge, skills, experience and past performance) of decision-makers. Every specific decision concerns a discussion matter. As a consequence, decision makers should master such matter to take suitable decisions. Thus, there exists a problem related to the context in which decisions are made.
The proposed decision support framework adopts fuzzy linguistic consensus model [[6], [7], [8]] enabling to reach consensus about the final decision considering decision-makers’ opinions conveniently expressed through linguistic values [[9], [10], [11], [12]] and their heterogeneous influence [[13], [14]]. The framework integrates a reinforcement learning algorithm [15] to incrementally update and consequently assign an appropriate weight (i.e., level of influence) to each decision-maker. The weights are assigned taking into account the context (i.e., the representation of the discussion mater) and how much time elapsed from the last involvement of each decision-maker. Ontologies and description logic are adopted to model and reason about contexts and to support weight assignment to each decision-maker. In our case study, we embed the weights of participants in the fuzzy decision-making paradigm to more effectively support the execution of the assembly phase in the value net. The proposed framework implements a weight learning algorithm that, by using context features, is able to rank the level of influence of decision-makers coherently to their competences within specific contexts. This will be proved by comparing the algorithm results to the roles of involved decision-makers. Every specific decision concerns a discussion matter. As a consequence, the
The proposed framework has been applied to footwear manufacturing company and illustrative examples are given by simulating decision-making activities within the assembly of the process of a value net in the domain of footwear manufacturing that is characterized by:(i) short life-cycles, because products are often designed to capture the current mood; (ii) high volatility, because the demand for these products is influenced by several and variable events (e.g., films, pop stars); (iii) low predictability, due to demand volatility that cannot be easily and/or accurately anticipated; (iv) high impulse purchasing, because consumers choose products when they see them directly in physical or online shops; and (v) high emotional appeal because consumers are looking for emotional and experiential elements in their purchasing choices. These factors lead to organizational and strategic changes in terms of specific supply chain strategies such as just-in-time [16], quick response systems [17] and agile supply chains [16]. In this scenario, many different core processes of footwear value chains are progressively outsourced, in other words, giving rise to complex and global value nets.
The proposed framework integrates several well assessed techniques in the area of context-aware computing and group decision making. The main novelty is in terms of application of these methods to face decision making problem in a complex organizational context. The framework is an extension of our previous research works early proposed in [[18], [19]], it relies on well-known fuzzy linguistic consensus model [[20], [21]]. The main contributions and distinguishing features with respect to previous research work are listed below:
- •
The framework has been instantiated to support decision-making in business process outsourcing (BPO) for a real industrial scenario applied to an Italian footwear company; it contributes to the knowledge advancement in the field of strategic business outsourcing, proposing a support model for managers who have to make a choice in a complex decision-making context.
- •
The case study revealed that the system could be used also to support human resource management measuring decision-making competences and relating them to specific skills of decision makers.
- •
The core framework adopts fuzzy linguistic variant of fuzzy consensus model instead of numerical approach used in [18], fuzzy linguistic approach is more suitable to be adopted in a concrete scenario because it provides linguistic terms to express decision-makers’ preferences.
- •
The knowledge based is time-aware, in fact, we introduced a 5-tuple (instead of 4-tuple of [18]) in which for each opinion given by a decision-maker there is a timestamp to trace the decisions. The time dimension has been used in the weights learning component in order to consider the elapsed time from the last involvement of decision maker a certain context, so the update of the weight value will depend on the elapsed time variable.
- •
The weight update induced by the contribution of a decision-maker in a given context (i.e., the discussion matter in which the decision must be taken) is used to update also the weights of decision maker for similar contexts (by using a propagation techniques fostered by the semantic technologies adopted to model and represent discussion matters) in order to keep the knowledge base coherent.
The paper is structured as follows: Section 2 analyses applications of group decision making (GDM) related to the proposed one; Section 3 provides a business process model (BPM) perspective of the investigated problem and introduces the overall approach; Section 4 details the proposed framework, namely context-aware heterogeneous group decision making (CaHGDM); then, Section 5 analyses the simulation results obtained by applying the proposed framework into an Italian footwear company. Finally, conclusions and future works are discussed.
Section snippets
Related works
The proposed framework supports decision-making for BPO when decision-makers with different background, skills, attitudes have to reach a convergence about possible alternative to select as final choice.
GDM is typically instantiated to face with Value Chain and Supply Chain, where the supporting decision-making is one of the most important requirement to more effectively and efficiently address, for instance, selection of supply chain partners, selection of vector system, selection of suppliers
Group decision making for business process outsourcing in value nets
This section provides a BPM perspective on the issues introduced in Section 1 and uses such perspective in order to describe, more in details, the overall approach in which the proposed framework is injected.
The core framework addressing GDM
The proposed framework is aimed at supporting group decision making problem in the context of business process outsourcing. We adopt the consensus model proposed in [[20], [21]] and its context-aware extension introduced in [18] to more effectively support group decision-making activities tanking advantage of previous experience of involved decision-makers in the same or similar context. More formally, given a set of decision-makers DM = {dm1, dm2, …, dmm}, (m ≥ 2) and a set of alternatives
Premise
This section shows the simulation results obtained by performing several executions of the group decision-making activity by means of the proposed CaHGDM framework for BPO in the context of an Italian footwear company. In particular, the company operates in the business from 50 years and represents one of the most important Italian companies active in the footwear industry with 46 million euro turnover, 300 employees and almost 300,000 pairs of shoes per-year, even if it is structured as a
Conclusions and future works
CaHGDM framework has been instantiated for an Italian footwear company to support decision-making in business process outsourcing. In particular, fuzzy linguistic consensus model is adopted for selecting the most suitable provider for outsourcing among the available ones. The proposed framework uses the context, in which decisions take place, and decision-makers are supposed to be managers with heterogeneous competence. The system implements a weight learning and assignment algorithm, that is
Acknowledgements
The authors are grateful to Fabi Spa (Italian footwear company) that provided the application scenario.
This work was supported by the FEDER funds under Grant TIN2016-75850-R.
References (56)
- et al.
Developing a decision model for business process outsourcing
Comput. Oper. Res.
(2007) - et al.
Building consensus in group decision making with an allocation of information granularity
Fuzzy Sets Syst.
(2014) - et al.
The interactive consensus reaching process with the minimum and uncertain cost in group decision making
Appl. Soft Comput.
(2017) - et al.
A review of soft consensus models in a fuzzy environment
Inform. Fusion
(2014) - et al.
Creating knowledge databases for storing and sharing people knowledge automatically using group decision making and fuzzy ontologies
Inform. Sci.
(2016) - et al.
A framework for context-aware heterogeneous group decision making in business processes
Knowl.-Based Syst.
(2016) - et al.
A model of consensus in group decision making under linguistic assessments
Fuzzy Sets Syst.
(1996) - et al.
A consensus based group decision making methodology for strategic selection problems of supply chain coordination
Eng. Appl. Artif. Intell.
(2013) - et al.
A linguistic mobile decision support system based on fuzzy ontology to facilitate knowledge mobilization
Decis. Support Syst.
(2016) - et al.
Personalized individual semantics in computing with words for supporting linguistic group decision making. An application on consensus reaching
Inform. Fusion
(2017)