Explainable predictive monitoring of temporal measures of business processes
Date
2019-01-07
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
Kaasaegsed ettevõtte infosüsteemid võimaldavad ettevõtetel koguda detailset informatsiooni äriprotsesside täitmiste kohta. Eelnev koos masinõppe meetoditega võimaldab kasutada andmejuhitavaid ja ennustatavaid lähenemisi äriprotsesside jõudluse jälgimiseks. Kasutades ennustuslike äriprotsesside jälgimise tehnikaid on võimalik jõudluse probleeme ennustada ning soovimatu tegurite mõju ennetavalt leevendada.
Tüüpilised küsimused, millega tegeleb ennustuslik protsesside jälgimine on “millal antud äriprotsess lõppeb?” või “mis on kõige tõenäolisem järgmine sündmus antud äriprotsessi jaoks?”. Suurim osa olemasolevatest lahendustest eelistavad täpsust selgitatavusele. Praktikas, selgitatavus on ennustatavate tehnikate tähtis tunnus. Ennustused, kas protsessi täitmine ebaõnnestub või selle täitmisel võivad tekkida raskused, pole piisavad. On oluline kasutajatele seletada, kuidas on selline ennustuse tulemus saavutatud ning mida saab teha soovimatu tulemuse ennetamiseks.
Töö pakub välja kaks meetodit ennustatavate mudelite konstrueerimiseks, mis võimaldavad jälgida äriprotsesse ning keskenduvad selgitatavusel. Seda saavutatakse ennustuse lahtivõtmisega elementaarosadeks. Näiteks, kui ennustatakse, et äriprotsessi lõpuni on jäänud aega 20 tundi, siis saame anda seletust, et see aeg on moodustatud kõikide seni käsitlemata tegevuste lõpetamiseks vajalikust ajast. Töös võrreldakse omavahel eelmainitud meetodeid, käsitledes äriprotsesse erinevatest valdkondadest. Hindamine toob esile erinevusi selgitatava ja täpsusele põhinevale lähenemiste vahel.
Töö teaduslik panus on ennustuslikuks protsesside jälgimiseks vabavaralise tööriista arendamine. Süsteemi nimeks on Nirdizati ning see süsteem võimaldab treenida ennustuslike masinõppe mudeleid, kasutades nii töös kirjeldatud meetodeid kui ka kolmanda osapoole meetodeid. Hiljem saab treenitud mudeleid kasutada hetkel käivate äriprotsesside tulemuste ennustamiseks, mis saab aidata kasutajaid reaalajas.
Modern enterprise systems collect detailed data about the execution of the business processes they support. The widespread availability of such data in companies, coupled with advances in machine learning, have led to the emergence of data-driven and predictive approaches to monitor the performance of business processes. By using such predictive process monitoring approaches, potential performance issues can be anticipated and proactively mitigated. Various approaches have been proposed to address typical predictive process monitoring questions, such as what is the most likely continuation of an ongoing process instance, or when it will finish. However, most existing approaches prioritize accuracy over explainability. Yet in practice, explainability is a critical property of predictive methods. It is not enough to accurately predict that a running process instance will end up in an undesired outcome. It is also important for users to understand why this prediction is made and what can be done to prevent this undesired outcome. This thesis proposes two methods to build predictive models to monitor business processes in an explainable manner. This is achieved by decomposing a prediction into its elementary components. For example, to explain that the remaining execution time of a process execution is predicted to be 20 hours, we decompose this prediction into the predicted execution time of each activity that has not yet been executed. We evaluate the proposed methods against each other and various state-of-the-art baselines using a range of business processes from multiple domains. The evaluation reaffirms a fundamental trade-off between explainability and accuracy of predictions. The research contributions of the thesis have been consolidated into an open-source tool for predictive business process monitoring, namely Nirdizati. It can be used to train predictive models using the methods described in this thesis, as well as third-party methods. These models are then used to make predictions for ongoing process instances; thus, the tool can also support users at runtime.
Modern enterprise systems collect detailed data about the execution of the business processes they support. The widespread availability of such data in companies, coupled with advances in machine learning, have led to the emergence of data-driven and predictive approaches to monitor the performance of business processes. By using such predictive process monitoring approaches, potential performance issues can be anticipated and proactively mitigated. Various approaches have been proposed to address typical predictive process monitoring questions, such as what is the most likely continuation of an ongoing process instance, or when it will finish. However, most existing approaches prioritize accuracy over explainability. Yet in practice, explainability is a critical property of predictive methods. It is not enough to accurately predict that a running process instance will end up in an undesired outcome. It is also important for users to understand why this prediction is made and what can be done to prevent this undesired outcome. This thesis proposes two methods to build predictive models to monitor business processes in an explainable manner. This is achieved by decomposing a prediction into its elementary components. For example, to explain that the remaining execution time of a process execution is predicted to be 20 hours, we decompose this prediction into the predicted execution time of each activity that has not yet been executed. We evaluate the proposed methods against each other and various state-of-the-art baselines using a range of business processes from multiple domains. The evaluation reaffirms a fundamental trade-off between explainability and accuracy of predictions. The research contributions of the thesis have been consolidated into an open-source tool for predictive business process monitoring, namely Nirdizati. It can be used to train predictive models using the methods described in this thesis, as well as third-party methods. These models are then used to make predictions for ongoing process instances; thus, the tool can also support users at runtime.
Description
Keywords
business processes, process management, business process modeling