حمایت از مدیریت تعارض در طراحی مشترک: یک روش برای ارزیابی اثرات تغییر مهندسی

It is characteristic of collaborative engineering design that precedence relationships among design activities contain information flow conflicts. Due to multi-actors interaction, conflicts can emerge from disagreements between designers about proposed designs. Hence, a critical element of collaborative design is to manage the detected conflicts and particularly the impacts once they are resolved. Indeed, the conflict resolution comes up with a solution which often implies modifications on the product and the process organisation. This paper deals with the problem of conflict management process and particularly the impacts of changes once a conflict is resolved. First, a solution is proposed to assess the impact on the product data based on the concept of data dependencies network. Second, this paper quantifies key issues with regards to Concurrent Engineering that enables us to better manage the design process. Strategies to overlap coupled activities are proposed based on the dependencies between the handled data during the design process. Furthermore, prospects to effectively re-organise the execution of design activities are proposed. Key features for a better process re-organization are studied: overlapping rate and multifunctional interaction rate. The author examines, under varying uncertainty conditions, how these key parameters affect the process performance of product development time and effort. Findings and recommendations to optimise the process re-organisation are summarized.

Although in most design processes, coordination entails clear communication between designers, the real reason for this coordination is not for communication but for resolving dependencies between product data [1]. Design is constraint oriented, and comprises many interdependent parts. Designers must consider not only the functional requirements of a product but other aspects such as geometrical, behavioural and structural, among others. Each of these has its own set of constraints which may contain conflicting or unsatisfied requirements and designers cannot always oversee the various alternatives and constraints. Then, due to the multi-actors interaction, conflicts emerge from disagreements between designers about incompatible and interdependent proposals. A critical element of collaborative design would be the conflict management, which can be perceived as the succession of mainly four phases: conflict detection, identification of the conflict resolution team, conflict resolution, and solution impact assessment. Several researchers have investigated conflict management in collaborative design, for example, Klein [2] and [3], Matta et al. [4], Cooper and Taleb-Bendiab [5], Lu et al. [6], Lara and Nof [7], and Rose et al. [8]. Most have proposed methods to support the conflict detection and the conflict resolution phases mainly. In a previous work [9], a solution called DEPNET1 has been developed to address the problem of identifying actors to be involved in the negotiation process to resolve the conflict. However, the assessment of the selected solution impact, once the conflict is resolved, has not yet been tackled in the previous work. Consequently, the aim of this paper is to extend DEPNET in order to assess the impact of a selected solution on the product as well as on the design process organization. The DEPNET solution is based on a process traceability system which explicitly captures and qualifies product data dependencies and inserts them in a dependencies network that is maintained throughout the design process. Based on this network, the conflict resolution team is identified to resolve this conflict. Indeed, conflict resolution cannot be achieved by one single actor since it requires the interplay of different areas of expertise. To avoid iteration in the conflict resolution process, it is highly advisable to do it in a collaborative way that seeks the input of many actors to reach a consensus quickly. These actors refer to those designers producing the product data leading to the source of the conflict. In order to identify these negotiators, the data dependencies network is extracted and the product data, on which the source of conflict depends, are identified through network backtracking (the source of conflict being the starting point). A set of queries are then applied in order to identify the negotiators forming the conflict resolution team. Once the team is identified, they collaborate during a negotiation phase in order to come up with a solution to the detected conflict. The negotiation leads to a solution which often implies changing one or more input data of the activity where the conflict has emerged, and thus, generating a cascade of modifications on the already produced data. The impact propagation is therefore highly dependent on the handled data during the design process. It supposes knowing the dependency relationships between the conflict source data and the data previously produced. Thus, it arises that propagating the selected solution impact mainly consists of identifying the dependency relationships between engineering data handled during the design process. Furthermore, the impacted data have to be redefined. This requires a re-execution of the design activities responsible on the elaboration of these product data and also an adjustment to the preliminary design process organisation. In addition to the introduction, this paper consists of five more sections. Section 2 presents a literature review of previous studies on product data dependencies and product development organization relevant to the described research work. Section 3 illustrates the problem definition, using the example of a turbocharger design process. Section 4 focuses on the data dependencies network through describing the network nodes (product data) and network arcs (data dependencies). Section 5 illustrates, by means of a case study, the use of the DEPNET method during conflict management focusing on solution impact on product data. Sections 6 and 7 present a framework to first assess the solution impact on process organization, and second to study key parameters to optimize the process rescheduling. Finally, Section 8 summarizes the paper and discusses future research opportunities.

Scheduling and managing emerging activities during collaborative design process is a difficult exercise that requires much effort and experience. Such emerging activities could occur following from conflict resolution. An examination of current process organisation scheduling tools reveals the weakness in various areas, such as identifying, sequencing, monitoring and controlling the emerging activities during the design process. In this paper, a process organisation framework based on the DEPNET method is presented to tackle the problems encountered in scheduling the complex and interrelated tasks for design process. Based on such a prototype framework, a complex collaborative design process can be monitored and controlled effectively. A simple design process rescheduling example was used to demonstrate the function of the rescheduling techniques proposed. The results were promising and revealed the potential of the framework. Nevertheless, further points remain to be considered on the issue of process reorganisation management. The results presented in Fig. 8a and b were obtained following from analyses carried out starting from the results of simulations and observations obtained in [27] and [38], and starting from a study of the relationship between the various uncertainty condition attributes and applied to the case of the turbocharger process. The overlapping and interaction rates, shown in Fig. 8a and b, are as an indication and represent tendencies rather than exact values. These indications were conducted and analysed with the design engineers of the industrial partner; and then discussed with the authors of the reference [27]. The results and conclusions presented in Section 7.2 are based on real case studies. The author studied several ongoing projects during several months (focusing on the turbine and impeller design phases) and how the overlapping techniques (overlapping rate and multifunctional rate) influence the rescheduling of the turbocharger design process. Hence, future work is to develop a computer-based model to simulate a design process, which will contain effort, development time, and uncertainty conditions. The aim of such model would be, first to simulate the whole design process taking into account customers as well as suppliers constraints. Second, rather than fixing the overlapping and interaction rates to four different levels (cf. Section 7.1); the main objective is to foresee the optimal values of these parameters. This is essential to validate and generalize the proposed framework. On the other hand, future work will include assessing the importance of each product data input which could be done by allocating different weights to each of them. This would be very useful to coordinate the data exchange in the case where an activity has several inputs from different activities. The project manager will reschedule the process accordingly. Such concept was introduced for example by Chen et al. [43] and Vajna et al. [44] to enhance process parallelization.