Organization, setting-up and management of projects

A project starts with the intention of producing an innovation (a new entity, a new product, service, and structure), induced by one or several ideas considered opportune. Such intention is then transformed into an organized work to obtain a tangible result, given to a team who has got the skills required and who tries to reach the aims of this result which have been set up simultaneously to the intention (in terms of time-limit, cost and quality). Such objectives enable to judge of the quality of realization of the project. Organizations commonly use this notion of project, and the concept conveyed appears in a variety of forms according to the nature of the result expected. The field of investigation is very broad: technical project of definition of a new product or of a new service close to the preoccupations of researchers from all fields of research, but also project of reorganization of a structure (remelting of the services, merger of companies, local development?) which are still in the sphere of activity of the engineer in industrial engineering. One can barely use standard process, besides, current project management turns towards the definition of processes adjusted to generic categories of projects. Researchers from the Industrial Engineering department focus on creating new products, in particular in the field of health, the information systems projects and cooperative ones.

Project and risk management

(D. Gourc, H. Pingaud, F. Benaben)

Project management aims to define and apply methodological principles to organize, pilot and manage projects. The organization goes through a work breakdown, a selection of team members, a mobilization of resources by scheduling, of modes of communication within the organization, of shared responsibilities etc. When all those elements are organized, the execution of phase can then start. Project management then deals with piloting and synchronization of operational processes in order to meet the cost, quality and time requirements. An efficient piloting requires a project instrumentation: setting-up of relevant project indicators, project report sheets. Our experience confirmed that the three aspects (organization, piloting and instrumentation) were indissociable, and that the team should work with deep consciousness of the relationship between those three parts. A change on one of the three sub-themes may necessitate to challenge experience on other themes. This ?waterfall effect? is very rich on the methodological level. The choice of a new instrumentation, for instance, will imply to redefine the communication mode among the team. Or still, the advocated piloting will introduce new tools for the project manager; hence the title of the whole research work.

However, most projects derive with the appearance of events, generally unfavourable ones. Since 1999, the department developed research work on identifying and taking into account uncertainty and risks (or opportunities) in projects. In 2004, we took part in three studies with industrials of the health sector and the life sciences interested by the implementation of such knowledge on their management problem. These studies were carried out in collaboration with the consulting firm MANAGEOS. This firm, created and managed by Sophie Bougaret, a former PhD student in the department, gives a very concrete continuation to her thesis? results. Connecting researchers and industry on real needs creates a ?win ? win? dynamic for both parties. Consequently, the researcher simultaneously validates his concepts and identifies new fields of research whilst industry enriches its practices.

In parallel, and still during the first half of the year, a collective thought was led on identifying risks in projects. Indeed, experience feedback on the field highlighted the fact that risk identification systematically is the weakest phase of the putting into practice of our way of conceiving management. A process of identifying risks was imagined, enabling to associate the practical knowledge of a professional tradesman to the complementary vision of a risk specialist. In this manner, The RIR approach identifies the specific risks of the project studied in confronting a typology of generic risks to the project activities.

Project management of information systems

(D. Gourc, H. Pingaud, F. Benaben, B. Ravalison, F. Darras, P. Gaborit)

ERP projects are particular ones of information systems, based on the purchase of a software package. Those software packages are well known for making MOA/MOE dialogs difficult. In the thesis work of F. Darras, we assume that the customer has already chosen his ERP solution. Our study focuses on the adjusting phase of a software package of the market to the company needs.

First, our approach aims to model the operational processes of the company, regardless of the software package chosen and secondly, to set up a relation between this model and the generic one of this ERP. We proposed a meta-model of the company to lay down the framework of the dialogue, then UML diagrams to cover the various viewpoints of the company needs (functional, informational, organizational and resources). Besides, from the dimensions of abstraction and genericity of the framework of the standard CEN ENV 40 003, we showed that the project management was corresponding to a transition between a variety of models at a conceptual level. The needs analysis, then the parametrizing of the application are consequently based on a specification in UML language of the customer?s need on the one hand, fitting the capacities of the application, in UML language too, on the other hand. The unicity of language guarantees a possibility of setting up a relation that is currently lacking to the people in charge of the integration. For a practical result, we worked for the account of an ERP editor who tries to provide his customers with this type of services to improve the use and penetration of his products. The thesis defense of Franck Darras took place in October 2004. The report is confidential. Further work is in progress by the editor, in relation with the research department.

In the framework of owner assistance on information system projects, the thesis of Benalisoa Ravalison focuses on the phases of identification and quantification of risks. Such works led to the collaboration between the consulting firm Comoe Toulouse, the Ecole des Mines of Alès (Mrs. B. Vacher) and with the Observatoire Régional de la Santé. After establishing a system of reference of risk sources, the student proposes a guidebook to structure the interviews to be led and make a list of the information about risks that have to be collected.

His work naturally inherits the ideas of the RIR approach, as mentioned above. From a generic list of risk factors and risks, the information collected can be transformed into a structured list of potential risks of the project, by simple comparative reasoning. This is a mode of rather deductive reasoning. But in order to facilitate and then to back up his diagnosis, the student examined in parallel a more inductive process, based on the theory of the agencement organisationnel. The point is to synthesize the knowledge collected on the project by the connections observed between the resources of the ground (human, symbolic, material ones, software package?). The construction of graphs representing the organizations enables to begin a reasoning over the validity of the organization in its state observed. The reasoning is then deduced in relation to the respect of rules or principles of project management to enable the critical points that feed risk analysis to emerge. Benalisoa studied two grounds in 2004; one with Comoe, one with the Observatoire Régional de la Santé. His thesis defense took place on February 1^{st, 2006.}

However, they are considered as taking part in the activities worth a major economic value. Our approach in that field aims to build up a rich communication in enhancing the value of the model to support the complexity of the study. The work on defining the scope of the study, of knowledge of management processes and of parametrizing of the application is based on a specification in UML language of the customer?s need on the one hand, and of the capacities of the application on the other hand. For a practical result, we worked for the account of an ERP editor, within the framework of the thesis of F. Darras. This editor tries to provide his customers with this type of services to improve the use and penetration of his products.

The project management and the project control set up a framework to the organization of the work in which one respects the rules of functioning, increasing the chances of meeting the objectives. Such rules induce a work breakdown, a choice of team members, a mobilization of resources thanks to scheduling, modes of communication within the team, shared responsibilities? Nowadays, the generalized use of such a framework in organizations mobilizes a broad and pluridisciplinary scientific community who observes the validity of rules and analyses the experience feedbacks. Such knowledge is taken over by a progressive evolution of the framework and rules in order to display efficient solutions for the projects that are considered the more complex ones. The researchers of the Industrial Engineering department who work on that topic are directly concerned by this process of constant analyze and adaptation. As a whole, the works contribute to enrich the knowledge of the methods of project management adapted to each category of project (R&D project, organizational project?). In that way, our contribution aims to propose theoretical and practical solutions on a certain number of cases, always connected to the industry. The rules of functioning applied without proper judgement lead to very strict processes harming the quality of the project?s progress as soon as some unexpected events occur and disturb it. We can quote some classic examples: an optimistic planning would be made obsolete by the delay of a task on the critical path. Likewise, an incomplete analysis of the project might not identify some tasks that should be taken into account to achieve the project, a lack of synchronizing of the information could induce a delay in the activity schedule, any change in the leader team might lead to modifying the project?s scope and objectives and would question the work schedule, the disfunctioning of a supplier would delay a planned delivery.

According to us, the identification and taking into account of risks in the definition of the project, their supervision during the course of the project enable to ease the rules of functioning of the project team and to progressively adapt the course of the project in relation with the evolution of the context and with its own evolution. The first series of our research works are directly connected to this risk management for a better project management. The thesis of S. Bakir, whose defense took place in June 2003, provided us with generic results, as the integration of processes of risk management and production planning of project have been explained in it in a clear and complete way. This scientific maturity has been turned to good account on two types of tricky projects:

• the projects of information systems, as dealt with by the thesis of B. Ravalison. In the framework of an owner assistance, he spent the first year working on the phases of identification and quantification of the risks induced by such initiatives within the organizations. Such works led to the collaboration between the consulting firm Comoe Toulouse and the Ecole des Mines of Alès (B. Vacher).

Management of collaborative projects

(D. Gourc, H. Pingaud, F. Benaben, L. Dupont, P. Gaborit, C. Villarreal, J. Touzi)

Collaborative projects within groups of SMEs concern the work of C. Villarreal. From the results on organizational trajectories provided by the OMSI group of the Ecole des Mines of Saint-Etienne, she tried to define a mode of management at the operational and tactical levels for this type of network organization. The sharing of means and the distributed nature of decision-making are two strong characteristics of that type of organization, which legitimate risk management. A bibliographic study on the theory of cooperative games enabled to reinforce this viewpoint. But such a management requires the improvement of the integration model as proposed by S. Bakir in a previous PhD thesis. C. Villarreal consequently worked on two levels. She adjusted risk analysis to this type of projects. The RIR approach has thus been tested. In parallel, she focused on the connections between the group?s management and the management of one (or several) project(s).

A study of the regional mechanical trade association of Midi-Pyrenees, constituting a group, enabled to feed her theories and to confront a domain very difficult to delimit. This study has been carried on in collaboration with the platform of studies and services on enterprise networks of the school and the ?Laboratoire de Gestion et de Cognition? of the UPS. Her thesis defense took place in 2005.

Reality of enterprise cooperation nowadays goes through connecting their information systems. Most of the time, data have to be exchanged and activities forming a collaborative process connecting partners synchronized. Process modelling is used to analyze the need and induce a work of specification that would make the applications composing the information systems interoperable. Jihed TOUZI joined our team in September 2004. His work stands at a conceptual level and deals with the connection of two information systems of two partner companies, whether it is in a supply chain or in a company network. He takes over the work from Franck DARRAS on information system projects in focusing on their cooperative feature. The aim is to specify, then implement a translator enabling to switch from the modelling of a collaborative process in BPMN language to a representation of the information system of partners in UML. The translator is hence based on the knowledge of both partner?s systems and takes into account their connecting under all viewpoints. This work was carried out in collaboration with EBM WebSourcing, a young company whose aim is to offer solutions to assure the interoperability between companies from a component-based approach. The result of the translator would then be used to facilitate the selecting work, then the component assembling one to meet the customer?s needs. At a project management level, this form of work gives adaptability to the redefinition of processes, the adjustment between the strategy of partners and functionalities of the cooperative information system is facilitated, or even partly automated if workflow tools are being incorporated to the system.

Such positioning allowed the team to position itself on the European InterOp excellence network in 2005, and to take part to the setting-up of the RNTL project (?Réseau National des Technologies Logicielles?): the JonES project (for Java Open ESB). The partners concerned are: EBM WebSourcing, France Telecom R&D, Inria ObjectWeb, Inria Sardes, Inria Jacquard, Odonata, OpenWide, ScalAgent.