Sünteetiliste sündmuste logide genereerimine baseerudes mitmeperspektiivsetele ärireeglitele
Files
Date
2016
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
Traditsiooniline äriprotsesside modelleerimine kasutab imperatiivset lähenemist, kus äri-protsesse kirjeldatakse üksteise järel sooritatavate tegevuste abil. On näidatud, et imper-atiivne lähenemine on sobivam lahendus stabiilsete ja ennustatavate protsesside puhul. Deklaratiivsed mudelid seevastu sobivad muutuvate protsesside kirjeldamiseks. Deklaratiivne mudel sisaldab endas reeglite hulka mida ei tohi eirata protsessi käitamisel. Viimastel aastatel on arendatud mitmeid uusi meetodeid deklaratiivsete protsessimudelite leidmiseks sündmuste logidest. Meetodite testimiseks on vajalik tööriistade olemasolu, mis genereerivad sünteetilisi sündmuste logisid, mille peal neid meetodeid katsetada. Enamus olemasolevaid tööriistu kasutavad imperatiivseid protsessimudelid logide genereerimiseks. Selline lähenemine ei ole sobiv deklaratiivsete protsessimudelite avastamise meetodite tes-timiseks. Sarnaselt on olemas vajadus tööriistade järgi, mis genereeriks sündmuste logisid kasutades mitmeperspektiivseid Declare mudeleid. Käesolevas töös esitleme tööriista mitmeperspektiivsete Declare mudelite genereerimiseks. See töörist tõlgib Declare piirangud lõpliku olekumasina esitusse,et neid kasutada deklaratiivsete mudelite simu-leerimiseks. Tööriist võimaldab kasutajatel genereerida logisid eeldefineeritud omadustega ( näiteks protsessi instantside arv ja protsessi pikkus), mis on kooskõlas Declare mudeli-tega.\n\rMärksõnad: Declare, deklaratiivne protsessimudel, protsessi simuleerimine, logide gene-reerimine, mitmeperspektiive, lineaarne taisarvuline planeerimine
Traditional business modelling is imperative in the sense that activities are provided step by step, from start to end, leading towards full business process. It has been proved that the imperative paradigm is most suitable in the context of stable and predictable processes. Declarative models are more suitable for variable processes. A declarative model is made of a set of constrains that cannot be violated during the process execution. In recent years, many techniques have been developed to discover declarative process model from event logs. To test these techniques it is sometime necessary to have tools that generate synthetic logs on which the techniques can be applied. However, majority of the existing tools avail-able in this field use simulation of an imperative process model to generate synthetic event logs. These approaches are not suitable for the evaluation of process discovery techniques using declarative process models. Additionally, there is a need for tools to generate event logs based on the simulation of multi-perspective declarative models. To close this gap, we developed a tool for log generation based on multi- perspective Declare models. This mod-el simulator will base on the translation of Declare constraints into Finite State Automata for the simulation of declarative processes. The tool will allows users to generate logs with predefined characteristics (e.g., number and length of the process instances), which is compliant with a given Declare model.\n\rKeywords: Declare, Declarative Process Models, Process Simulation, Log Generation, Multi-perspective, Integer Linear Programming
Traditional business modelling is imperative in the sense that activities are provided step by step, from start to end, leading towards full business process. It has been proved that the imperative paradigm is most suitable in the context of stable and predictable processes. Declarative models are more suitable for variable processes. A declarative model is made of a set of constrains that cannot be violated during the process execution. In recent years, many techniques have been developed to discover declarative process model from event logs. To test these techniques it is sometime necessary to have tools that generate synthetic logs on which the techniques can be applied. However, majority of the existing tools avail-able in this field use simulation of an imperative process model to generate synthetic event logs. These approaches are not suitable for the evaluation of process discovery techniques using declarative process models. Additionally, there is a need for tools to generate event logs based on the simulation of multi-perspective declarative models. To close this gap, we developed a tool for log generation based on multi- perspective Declare models. This mod-el simulator will base on the translation of Declare constraints into Finite State Automata for the simulation of declarative processes. The tool will allows users to generate logs with predefined characteristics (e.g., number and length of the process instances), which is compliant with a given Declare model.\n\rKeywords: Declare, Declarative Process Models, Process Simulation, Log Generation, Multi-perspective, Integer Linear Programming