An Architecture for Service Modelling in Servitization Context: MDSEA1

Yves Ducq

IMS, UMR 5218, University of Bordeaux
351 Cours de la libération, 33405 Talence Cedex, France
yves.ducq@ims-bordeaux.fr

ABSTRACT: Currently, manufacturing enterprises are progressively migrating from traditional product-centric business to product-based service-oriented virtual enterprise. This paper aims at presenting the service modelling architecture that is proposed in the frame of a FP7 project: MSEE. The proposed Model Driven Service Engineering (MDSE) architecture is adapted from MDA developed by OMG and from MDI model developed in INTEROP Network of Excellence. This architecture defines the various modelling levels and the related constructs to model based on servitization principles. The proposed modelling languages to represent these constructs at each level of MDSE will be presented at the same time. Templates to describe each concept will be also illustrated in example. Conclusions and perspectives are given in the end of the paper.

KEYWORDS: service system modelling, enterprise modelling, MDI/MDA approaches

1. Introduction

In the economic context moving continuously, manufacturing enterprise, including SME’s, are progressively migrating from traditional product-centric business to product-based service-oriented virtual enterprise and ecosystems [1]. In this sense, traditional companies have to cooperate in one or several virtual enterprises, considered as service systems dedicated to support the service design, development, and implementation. In order to facilitate this migration process, service system that will provide desired services around the product will have to be modelled, designed, implemented, tested and managed along its entire lifecycle in order to ensure the correct migration of the company.

This paper presents a second version of a model driven architecture with associated modelling formalisms which are the results of a research work performed in the frame of the FP7 MSEE Integrated Project [2]. Particularly, the goal of this work is to develop service system modelling language to support service system engineering and implementation. The approach adopted is to use Enterprise Modelling techniques as a basis under the Model Driven Service Engineering Architecture extended from Model Driven Architecture developed by OMG and Model Driven Interoperability approach developed in INTEROP Network of Excellence [3].

So, in a first part, the servitization process and the definition of what a service system is will be developed as well principles of system modelling. Then, the Model Driven Service Engineering Architecture will be presented insisting on its interest for the implementation of a coherent and complete virtual enterprise based on business models and on the description of each modelling level. The various levels of the architecture will be detailed as well as the required constructs to represent at each level based on the previous servitization principles. Then, enterprise modelling languages will be proposed to represent the constructs at each level. Finally, the perspectives of this work will be proposed.

2. Servitization and virtual organization as a service system

The studies and researches in the domain of Service have been mostly devoted to support tertiary sector domains (e.g. banking & finance, tourism, trade, public administration), with an obvious focus on ICT. At the end of the nineties, the concept of Service in Manufacturing appeared and the evolution from an economy of products towards an economy of services round the products becomes more and more important in manufacturing: this evolution is called Product Service System (PSS) or Servitization.

According to Wikipedia “a service is to make available a technical or intellectual capacity” or “to supply a work which will be useful for the user without material transformation”.

Most of the time a service is opposed to a good. The following list characterizes a service [4]:

Since a decade, new research thinking has been emerging, trying to systematize the multi-disciplinary knowledge involved in service systems. On their web page, IBM describes service science as “a growing multi-disciplinary research and academic effort that integrates aspects of established fields like computer science, operations research, engineering, management sciences, business strategy, social and cognitive sciences, and legal sciences” [5].

In the computer science domain, Service Oriented Architectures (SOA), have revolutionized information systems, by providing software engineers with powerful methodologies and tools for decomposing complex systems into autonomous components. The final aim of such evolution is to support enterprise vital processes and workflows, by simple orchestrations and compositions in the hand of business specialists.

Clearly the servitization of manufacturing companies covers different levels of service provision and consequently different stages can be followed to evolve.

In servitization, the product is considered as the core element of the service to deliver to customers and subsequently we follow a manufacturing approach taking into account the market pressure that oblige to create new models in order to meet the servitization challenge. An appropriate concept to link products, product related services and the needs of the users is the “Extended Product” (EP) [6].

The Extended Product concept belongs to the category of Product-Service System. The Extended Product-Service is characterized by a layer model based on manufacturing product and defining the process extensions. The Extended Product is a complex result of tangible and intangible components.

The Core Product is the physical product that is offered to the market; while the Product Shell describes the tangible “packaging” of the product (e.g. one enterprise sells machine-tools and will add the maintenance which can be done by another company). Supporting Services are intangible additions, which facilitate the use of the product (e.g. maintenance plans or mobility guarantees). Differentiating Services provide individualization of the Extended Product on the market.

Then product extensions are described by the tangible and intangible aspects of a “utility package” to satisfy the customers’ needs. They can be used to gain competitive advantage by offering added value to the customer. While in the past production costs, marketing, quality and reliability or time to market have been key success factors, nowadays innovation is the decisive characteristic [7].

The resulting Extended Product would be the specific solution satisfying the customers demand. As the solution can become very complex, several business partners may be collaborating for the provision of the EP in the frame of an Ecosystem.

Customers are looking for solutions and benefits (not only to acquire products) or even more they are requesting intangibles like leadership on the market, success, fame, etc. Manufacturers need to package their core products with additional services to make them more attractive.

The different stages of service provision are shown in Figure 1.

Figure 1. From extended product concept to servitization process

The first stage is the selling of a product (e.g. a machine tool).

The second stage which initializes the servitization process and the evolution toward Product+Service, start by adding a simple service (Product and supporting service) (e.g. the company will add a device on the machine-tool allowing to check continuously the running of the machine etc.).

The third stage (Product and differentiating service) is an evolution of the previous one. The service is more elaborated and increases the differentiation (e.g. the company can propose to sell the machine plus a service which guaranties a high percentage of availability of this machine).

The fourth stage, Product2Service scenarios are in contrast sharply decoupling manufacturing of goods and selling of services, where in most cases physical goods remain the property of the manufacturer and are considered as investment, while revenues come uniquely from the services (e.g. the company does not sell the machine-tools but sells hours of running of the machine-tool).

However, in this context, this is difficult for most of the companies to work alone in the Product2Service scenario but need to be associated to other companies all along the service life cycle. Working together in design and production networks results in an extension of companies’ “ability to reach” e.g. wider markets and knowledge spheres, a higher level of the companies agility, as well as the possibility to share risks and resources. An industrial model for collaboration to exploit the various opportunities is a Virtual Manufacturing Enterprise (VME) (figure 2).

Figure 2. Virtual...