Suchergebnisse - "two complementary methods"

  • Treffer 1 - 3 von 3
Treffer weiter einschränken
  1. 1
  2. 2
  3. 3

    Autoren: Uba, Reina

    Schlagwörter: Companies that have years of experience in business process management often maintain repositories containing hundreds or even thousands of business process models. The models in these repositories usually originate from various sources and are developed by different stakeholders. A common practice is that new process models are created by extending or refining existing models, or by copying and merging fragments from multiple models. As a result, process models tend to accumulate duplicate fragments which, if left unconsolidated, may evolve independently and lead to inconsistencies. Also, it often occurs that organizations manage multiple business processes that have similar goals, but pertain to different customer types, different products, business units or geographical regions. For example, a business process for handling insurance claims for motor accidents shares the same goal as a business process for handling house insurance claims. Naturally, these models will share several common fragments, but will differ from one another at various points. Managing these processes as entirely separate entities, leads to redundancy and inefficiency. In this setting, this thesis addresses the following question: How to identify duplicate fragments in process model repositories, and more generally, how to identify and consolidate commonalities across models in a large process model repository? The thesis proposes two complementary methods for process model consolidation, namely process model merging and subprocess extraction. Process model merging takes as input two or more process models and produces a single consolidated model that analysts can use to manage entire families of similar process models rather than managing them independently. On the other hand, subprocess extraction is about identifying fragments that are shared by multiple process models (also known as clones) and encapsulating these clones as separate subprocesses in order to eliminate redundancies., business processes, models, software development, äriprotsessid, mudelid, tarkvaraarendus, dissertatsioonid, väitekiri, ETD, dissertation

    Dateibeschreibung: application/pdf

    Relation: Dissertationes mathematicae Universitatis Tartuensis;78; http://hdl.handle.net/10062/18709