Accurate and efficient discovery of process models from event logs
Date
2020-02-14
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
Töötajate igapäevatöö koosneb tegevustest, mille eesmärgiks on teenuste pakkumine või toodete valmistamine. Selliste tegevuste terviklikku jada nimetatakse protsessiks. Protsessi kvaliteet ja efektiivsus mõjutab otseselt kliendi kogemust – tema arvamust ja hinnangut teenusele või tootele. Kliendi kogemus on eduka ettevõtte arendamise oluline tegur, mis paneb ettevõtteid järjest rohkem pöörama tähelepanu oma protsesside kirjeldamisele, analüüsimisele ja parendamisele.
Protsesside kirjeldamisel kasutatakse tavaliselt visuaalseid vahendeid, sellisel kujul koostatud kirjeldust nimetatakse protsessimudeliks. Kuna mudeli koostaja ei suuda panna kirja kõike erandeid, mis võivad reaalses protsessis esineda, siis ei ole need mudelid paljudel juhtudel terviklikud. Samuti on probleemiks suur töömaht - inimese ajakulu protsessimudeli koostamisel on suur.
Protsessimudelite automaatne koostamine (protsessituvastus) võimaldab genereerida protsessimudeli toetudes tegevustega seotud andmetele. Protsessituvastus aitab meil vähendada protsessimudeli loomisele kuluvat aega ja samuti on tulemusena tekkiv mudel (võrreldes käsitsi tehtud mudeliga) kvaliteetsem. Protsessituvastuse tulemusel loodud mudeli kvaliteet sõltub nii algandmete kvaliteedist kui ka protsessituvastuse algoritmist.
Antud doktoritöös anname ülevaate erinevatest protsessituvastuse algoritmidest. Toome välja puudused ja pakume välja uue algoritmi Split Miner. Võrreldes olemasolevate algoritmidega on Splint Miner kiirem ja annab tulemuseks kvaliteetsema protsessimudeli. Samuti pakume välja uue lähenemise automaatselt koostatud protsessimudeli korrektsuse hindamiseks, mis on võrreldes olemasolevate meetoditega usaldusväärsem. Doktoritöö näitab, kuidas kasutada optimiseerimise algoritme protsessimudeli korrektsuse suurendamiseks.
Everyday, companies’ employees perform activities with the goal of providing services (or products) to their customers. A sequence of such activities is known as business process. The quality and the efficiency of a business process directly influence the customer experience. In a competitive business environment, achieving a great customer experience is fundamental to be a successful company. For this reason, companies are interested in identifying their business processes to analyse and improve them. To analyse and improve a business process, it is generally useful to first write it down in the form of a graphical representation, namely a business process model. Drawing such process models manually is time-consuming because of the time it takes to collect detailed information about the execution of the process. Also, manually drawn process models are often incomplete because it is difficult to uncover every possible execution path in the process via manual data collection. Automated process discovery allows business analysts to exploit process' execution data to automatically discover process models. Discovering high-quality process models is extremely important to reduce the time spent enhancing them and to avoid mistakes during process analysis. The quality of an automatically discovered process model depends on both the input data and the automated process discovery application that is used. In this thesis, we provide an overview of the available algorithms to perform automated process discovery. We identify deficiencies in existing algorithms, and we propose a new algorithm, called Split Miner, which is faster and consistently discovers more accurate process models than existing algorithms. We also propose a new approach to measure the accuracy of automatically discovered process models in a fine-grained manner, and we use this new measurement approach to optimize the accuracy of automatically discovered process models.
Everyday, companies’ employees perform activities with the goal of providing services (or products) to their customers. A sequence of such activities is known as business process. The quality and the efficiency of a business process directly influence the customer experience. In a competitive business environment, achieving a great customer experience is fundamental to be a successful company. For this reason, companies are interested in identifying their business processes to analyse and improve them. To analyse and improve a business process, it is generally useful to first write it down in the form of a graphical representation, namely a business process model. Drawing such process models manually is time-consuming because of the time it takes to collect detailed information about the execution of the process. Also, manually drawn process models are often incomplete because it is difficult to uncover every possible execution path in the process via manual data collection. Automated process discovery allows business analysts to exploit process' execution data to automatically discover process models. Discovering high-quality process models is extremely important to reduce the time spent enhancing them and to avoid mistakes during process analysis. The quality of an automatically discovered process model depends on both the input data and the automated process discovery application that is used. In this thesis, we provide an overview of the available algorithms to perform automated process discovery. We identify deficiencies in existing algorithms, and we propose a new algorithm, called Split Miner, which is faster and consistently discovers more accurate process models than existing algorithms. We also propose a new approach to measure the accuracy of automatically discovered process models in a fine-grained manner, and we use this new measurement approach to optimize the accuracy of automatically discovered process models.
Description
Keywords
business processes, business process modeling