| M. Voorhoeve and W.M.P. var den Aalst. Ad-hoc Workflow: problems and solutions. Proc. of the 8th DEXA Conference on Database and Expert Systems Applications, pages 36-41, Toulouse, France, 1997 |
....operations unlike a database wansaction; the nature of the business process could be forgiven to some errors thereby not requiting an undo operation. Therefore, the error handling semantics of traditional transactional processing systems are too rigid for workflow systems. Ad hoc workflows [28, 29] solve problems case by case. Our approach proposed in this paper support more than ad hoc workflows. In the first place, we use a context dependent approach to support adaptive exception handling. We support ad hoc workflows in the sense when exceptions occur, non standard exception handlers ....
M. Voorhoeve, W. Aalst, "Ad-hoc Workflow: Problems and Solutions", 36-40, DEXA Workshop 1997
....less complex, and often more sporadic processes for tasks such as expense form processing. They are easier to implement but also enforce rigid routing schemes. Ad hoc workflow systems support the definition of new, unanticipated work flows such as those needed to support collaborative work [33]. Ad hoc WFMS are close in spirit to the systems proposed here. As in our proposals, ad hoc WFMS are often implemented as enhancements to basic e mail transport mechanisms and they are often used to support information distribution. Our dynamic mailing list proposal may be classified as a hybrid ....
Voorhoeve, M. and Van der Aalst, W. "Ad-hoc workflow: problems and solutions." Proceedings of the 8th International Workshop on Database and Expert Systems Applications, Toulouse, France, 1-2 Sept. 1997, p. 36-40.
....respective step attributes) for the workflow. The advantage of our approach is that even this (meta )constraint may be violated with suitable excusing privileges. 6 RELATED WORK AND CONCLUSIONS To begin with, there have been many papers on the evolution of workflow and process models, including [1, 2, 4, 11, 18, 19, 26, 32, 34]. We believe there is a distinction to be made between deviations during workflow enactment (the topic of this paper) and workflow evolution. The difference is analogous to the one between allowing exceptional individuals in a database (e.g. 7] and schema evolution in a database (e.g. 5] ....
....from the tokens in the state of the original CICN to those of the modified CICN. In our examples (see Section 5.4) this mapping is essentially simulated by the creation of activity nodes for steps placed in the ENABLED class. The other Petrinet like approaches to modifying workflow models (e.g. [2, 34]) bear a similar relationship to our work. In fact, in [9] we show how the present framework can be applied to the workflow coordination model in [2] which has the nice theoretical property that one can analyze when it is safe to make changes. A second class of work, which usually deals with ....
M. Voorhoeve, W. van der Aalst. "Ad-hoc Workflow: Problems and Solutions", Proc. Workshop Databases and Expert Systems, 1997, pp.36-41.
....fully understand the concepts. At the moment, there are more than 200 workflow products commercially available [24] and many organisations are introducing workflow technology to support their business processes. It is widely recognised that workflow management systems should provide flexibility [7 9, 12, 16, 21, 28, 33, 36, 37, 47, 56]. However, today s workflow management systems have problems dealing with changes, e.g. new technology, new laws, and new market requirements may lead to (structural) modifications of the workflow process definition at hand. In addition, ad hoc changes may be necessary, e.g. because of ....
....this small example, it may seem trivial to obtain this information. However, in general there are many variants of processes which may have up to 100 tasks and it is far from trivial to present aggregated information to the manager. The topic of generating management information was addressed in [5 7, 55, 56]. Despite its relevance for the next generation of workflow management systems only few researchers seem to be working on this topic. E commerce refers to the enabling of purchasing and selling of goods and services through a communications network [32, 59] The ability to conduct business ....
M. Voorhoeve and W.M.P. van der Aalst. Ad-hoc Workflow: Problems and Solutions. In R. Wagner, editor, Proceedings of the 8th DEXA International Workshop on Database and Expert Systems Applications, pages 36--40, Toulouse, France, September 1997. IEEE Computer Society Press, Los Alamitos, California, 1997.
....ofworkflow processes. The addition of the horizontal dimension allows for the design and enactment of many variants of a workflow process. However, it is not sufficient to support the design and enactment. There are two additional issues that need to be dealt with: 1) management information [32,33], and (2) dynamic change [7,11,12] In spite of the existence of many variants of one process, the manager is interested in information at an aggregate level, i.e. management information which abstracts from small variations. The term dynamic change refers to the problem of handling old cases in ....
....transferred, a part that is handled the old way, and parts which are handled by hybrid process definitions (comparable to the approach using change regions) Reichert and Dadam[26] use a similar approach without addressing for example the problem identified in Figure 1. Voorhoeve and Van der Aalst [32,33] also propose a fixed set of transformation rules to support dynamic change. However, the drawback of using transformation rules is that only local changes are considered and the rules provided so far are far from being complete. Moreover, valuable information is lost during the application of a ....
M. Voorhoeve and W.M.P. van der Aalst. Ad-hoc Workflow: Problems and Solutions. In P Wagner, editor, Proceedings of the 8th DEXA Conference on Database and Expert Systems Applications, pages 36-41, Toulouse, France, Sept 1997.
....of workflow processes. The addition of the horizontal dimension allows for the design and enactment of many variants of a workflow process. However, it is not sufficient to support the design and enactment. There are two additional issues that need to be dealt with: 1) management information [49,50], and (2) dynamic change [13,20,21] In spite of the existence of many variants of one process, the manager is interested in information at an aggregate level, i.e. management information which abstracts from small variations. The term dynamic change refers to the problem of handling old cases in ....
....so called change regions , is proposed to avoid the anomaly illustrated by Figure 1. There has been some follow up work addressing this problem [ 13,21,34] In addition there are several papers on workflow change, not addressing and or avoiding the problem by a fixed set of transformation rules [17,27,30,36,41,49,50,52]. Independent of the approach used, the following two issues constitute a policy for dynamic change: o:o When to jump from the old process to the new process definition : Which state to jump to A good policy for the example shown in Figure 1 is the following. The right hand process will jump ....
[Article contains additional citation context not shown here]
M. Voorhoeve and W.M.P. van der Aalst. Ad-hoc Workflow: Problems and Solutions. In R. Wagner, editor, Proceedings of the 8th DEXA Conference on Database and Expert Systems Applications, pages 36-41, Toulouse, France, Sept 1997.
....approach using change regions) Reichert and Dadam (Reichert and Dadam, 1998) use a similar approach. However, semantical issues such as errors introduced by swapping tasks, skipping tasks, or multiple executions of a task are not considered. Voorhoeve and Van der Aalst (Voorhoeve and Aalst, 1996; Voorhoeve and Aalst, 1997) also propose a fixed set of transformation rules to support dynamic change. However, the rules are not given explicitly at the net level and semantical issues are not considered. Van der Aalst and Basten (Aalst and Basten, 2001) propose an approach based on inheritance. This approach uses a set ....
Voorhoeve, M. and Aalst, W. (1997). Ad-hoc Workflow: Problems and Solutions. In Wagner, R., editor, Proceedings of the 8th DEXA International Workshop on Database and Expert Systems Applications, pages 36--40, Toulouse, France. IEEE Computer Society Press, Los Alamitos, California, 1997.
....by hybrid process definitions (similar to the approach using change regions) Reichert and Dadam [52] use a similar approach. However, semantical issues such as errors introduced by swapping tasks, skipping tasks, or multiple executions of a task are not considered. Voorhoeve and Van der Aalst [60, 61] also propose a fixed set of transformation rules to support dynamic change. However, the rules are not given explicitly at the net level and semantical issues are not considered. The reader interested in workflow change and Petri nets is also referred to [6] which contains several papers of the ....
M. Voorhoeve and W.M.P. van der Aalst. Ad-hoc Workflow: Problems and Solutions. In R. Wagner, editor, Database and Expert Systems Applications, 8th. International Workshop, DEXA'97, Proceedings, pages 36--40, Toulouse, France, September 1997. IEEE Computer Society Press, Los Alamitos, California, 1997.
....by hybrid process definitions (similar to the approach using change regions) Reichert and Dadam [47] use a similar approach. However, semantical issues such as errors introduced by swapping tasks, skipping tasks, or multiple executions of a task are not considered. Voorhoeve and Van der Aalst [55, 56] also propose a fixed set of transformation rules to support dynamic change. However, the rules are not given explicitly at the net level and semantical issues are not considered. None of the papers mentioned uses an approach based on inheritance. The transfer rules based on the four ....
M. Voorhoeve and W.M.P. van der Aalst. Ad-hoc Workflow: Problems and Solutions. In R. Wagner, editor, Database and Expert Systems Applications, 8th. International Workshop, DEXA'97, Proceedings, pages 36--40, Toulouse, France, September 1997. IEEE Computer Society Press, Los Alamitos, California, 1997.
....of workflow processes. The addition of the horizontal dimension allows for the design and enactment of many variants of a workflow process. However, it is not sufficient to support the design and enactment. There are two additional issues that need to be dealt with: 1) management information [38,39], and (2) dynamic change [11,15,16] In spite of the existence of many variants of one process, the manager is interested in information at an aggregate 3 level, i.e. management information which abstracts from small variations. The term dynamic change refers to the problem of handling old ....
....transferred, a part that is handled the old way, and parts which are handled by hybrid process definitions (comparable to the approach using change regions) Reichert and Dadam[32] use a similar approach without addressing for example the problem identified in Figure 1. Voorhoeve and Van der Aalst [38,39] also propose a fixed set of transformation rules to support dynamic change. However, the drawback of using transformation rules is that only local changes are considered and the rules provided so far are far from being complete. Moreover, valuable information is lost during the application of a ....
M. Voorhoeve and W.M.P. van der Aalst. Ad-hoc Workflow: Problems and Solutions. In R. Wagner, editor, Proceedings of the 8th DEXA Conference on Database and Expert Systems Applications, pages 36-41, Toulouse, France, Sept 1997.
....trivial to obtain this information. However, in general there are many variants of processes which may have up to 100 tasks and it is far from trivial to present aggregated information to the manager. The topic of generating management information was addressed in (Voorhoeve Van der Aalst 1996; Voorhoeve Van der Aalst 1997). Despite its relevance for the next generation of workflow management systems only few researchers seem to be working on this topic. At the moment, only intuitive notions for gcd and lcm exist. 7. CONCLUSION In this paper, we discussed some of the problems that need to be solved to enable ....
Voorhoeve, M. & Aalst, W.M.P. van der 1997. Ad-hoc Workflow: Problems and Solutions. In R. Wagner, editor, Proceedings of the 8th DEXA Conference on Database and Expert Systems Applications, pages 36-41, Toulouse, France.
No context found.
M. Voorhoeve and W.M.P. var den Aalst. Ad-hoc Workflow: problems and solutions. Proc. of the 8th DEXA Conference on Database and Expert Systems Applications, pages 36-41, Toulouse, France, 1997
No context found.
Voorhoeve, M., van der Aalst, W. "Ad-hoc Workflow: Problems and Solutions", Proc. Workshop Databases and Expert Systems, 1997, pp.36-41.
Online articles have much greater impact More about CiteSeer.IST Add search form to your site Submit documents Feedback
CiteSeer.IST - Copyright Penn State and NEC