Catálogo de publicaciones - libros

This paper presents an approach to support collaborative modeling with graph based representations. In particular, the problem of partially shared models with associated semantics is addressed, and an architectural solution to enable flexible modes of partial application synchronization under the constraint of retaining a common semantics in the shared model parts is presented.

In fragmented work settings like network organizations or virtual organizations, monolithic approaches to implement support for collaboration would require the actors involved to agree on the usage of the approach or tool under consideration. As the autonomy of actors in these settings makes this hard to achieve, we suggest an exploration and an increase in the end-user tailorability of basic software infrastructures to enable even actors in theses settings to tailor their collaboration support to their needs. An example for this strategy is illustrated by using email as a basic groupware technology. We use server-based email filters to improve the coordination of work processes and increase group awareness in these settings, and focus on making it easy for end users to understand and tailor the technology according to their needs. We use and enhance concepts from the discussion on the "tailorability" of CSCW systems (a visual filter composition language, a component-based architecture and additional support for exploration and documentation) to implement and evaluate our prototype.

This paper proposes a collaborative framework handling unexpected exceptions in Workflow Management Systems (WfMS). Unexpected excep tions correspond to unpredicted situations for which the system can not suggest any solutions. We introduce the notion that exception recovery is a collabora tive problem solving activity that should be addressed through an intertwined play between several actors performing two types of tasks: (1) diagnosing situations; and (2) planning recovery actions. We propose a set of dimensions to classify the exceptional situations and their relations to recovery strategies. We also discuss the importance of monitoring recovery actions within the scope of diagnosis tasks. The proposed solution is implemented through a dedicated workflow.

This work presents a workflow process mining method that is, at least, as powerful as many others presented in the literature, as measured by the examples presented in the literature. The method is based on a grammar of rewriting expressions, by which a model is adapted to include a new execution trace. We also discuss the intrinsic limits of the mining process, which we believe has not been a topic clearly stated and discussed in the published research.

Languages that support object-oriented programming are now mainstream, and can support software reuse. This study focused on the reusability of components for workflow management systems (WfMSs). Implementing a WfMS in object-oriented programming languages without considering the characteristics of the WfMS does not ensure that all the components will be reusable. We first clarify the reusability of WfMSs and point out the difficulties in constructing reusable components for WfMSs. We then propose an object-oriented model for WfMSs named the “Java-based object-oriented WfMS” (JOO-WfMS), whose components are fine-grained and are classified into a functional stack with three layers. This extends the reusability of objects in developing workflow applications. The resulting architecture can support real-time flow control as well as the dynamic instantiation of objects. Two mechanisms are embedded into the JOO-WfMS to increase the reusability of its components: (1) a workflow failure-handling language, which can increase the reusability of activities when flexible failure recovery is necessary; and (2) the user communication components and their corresponding architecture. The goal of the architecture is to increase the reusability of codes used for communication between the user and the activities in WfMSs.

Using scheduling techniques to reorder work in workflow systems can improve the performance of the business processes enacted by these systems. But this research area is still under-explored. This paper presents preliminary results on ongoing research on the modeling of the behavior of scheduling techniques in workflow systems using statistical analysis. It discusses the the motivations for this approach and presents a first workflow scenario to be studied. Simulations of the use of dispatching rules to minimize the percentage of late cases are performed. The results are analyzed with statistical regression techniques to estimate functions that describe the observed data and compose a model to the behavior of the scheduling techniques. The model is evaluated against new data that was not present in the first observed data and the results show that it is feasible and could be used to assist systems administrators in applying scheduling techniques to workflow systems.

Organizations frequently need to recall past events that, for some reason, were not adequately documented when they occurred. The successful reconstitution of past events depends on several variables, such as how long ago the event occurred, and whether key people are still in the organization. It also depends on the supporting process. This paper examines three knowledge recall methods and compares them in a controlled experiment. A group storytelling approach is used in two of the methods, one of which is supported by technology. The results obtained favor the group approach, but the advantages of technology support are not conclusive. The paper also evaluates the benefits and the drawbacks of using a supporting technology.

Eliciting and re-using knowledge within an organization requires a very structured communication process among its employees in order to avoid misunderstanding and confusion. The transfer of knowledge among actors can only be successful if a common interpretative focus and its context are set up. So far, information about the real context that surrounded team’s past activities can help their members to better understand situations at hand. In this paper, we argue that a combination of group storytelling technique and a groupware tool can help the elicitation and use of the context shared by a group. Moreover, our main goal is to discuss how groupware can help to structure and formalize the contextual information behind the scenes of a story told by a group, making it easier to understand, interpret and reuse the knowledge intrinsic to it.

This work present an empirical evaluation of factors that discourage a stronger collaboration of physicians across time. By observing two different outpatient clinics in which a single patient is treated by a sequence of physicians for a long period, we were able to detect three aspects of a (paper) patient record that makes collaboration difficult: lack of diagnostic rationale, lack of treatment rationale and improper way of presenting the information contained in the patient record.

The knowledge building can be seen as a collaborative process of which negotiation is a fundamental aspect. The use of technology to support this process has been attempted in various groupware systems. However, there is no adequate support system for the process of negotiation, which generally relies on voting as the element for reaching agreement in decision-making. On the other hand, the best approaches to this problem have been formalised in learning environments where there is a clear hierarchical structure. When the environment is not hierarchical, new problems arise which require special attention. This article presents a model of knowledge building that has negotiation as its basis in a group that is non-hierarchical in its structure. The model is implemented on a prototype tool named ShaKnoMa, which is tested on common tasks in an environment such as that proposed here. For the knowledge representation, concept maps are used which act a scaffolding to classify, index and search the information.