Powered by OpenAIRE graph
Found an issue? Give us feedback
DBLParrow_drop_down
DBLP
Doctoral thesis
Data sources: DBLP
Comenius-Institut
Thesis . 2021
License: CC BY NC ND
Data sources: Comenius-Institut
versions View all 3 versions
addClaim

Robotic Process Mining: Accelerating the Adoption of Robotic Process Automation.

Authors: Leno, Volodymyr;

Robotic Process Mining: Accelerating the Adoption of Robotic Process Automation.

Abstract

Protsessi muutmine võib ettevõtetele osutuda kulukaks ja riskantseks, kuid vajalikuks. Muutuste eiramine võib avaldada mõju ettevõtte resurssidele, selle keskkonnale või jätkusuutlikusele. Üks ettevõtete poolt enimkasutatumaid meetmeid disainimiseks ja hindamiseks protsessi on äriprotsessi simulatsioon. See tehnika võimaldab luua hüpoteetilisi stsenaariume ja hinnata teostuse tagajärgi virtuaalses keskkonnas võtmata riski ebaõnnestuda reaalsuses. Modifitseerides protsessi üksikasju simulaatoris annab võimaluse analüütikutele teha oletusi nagu näiteks „kui sa eemaldad selle, juhtub see või kui sa lisad selle, siis võib juhtuda see.“ Selline võime on väga mugav abistamaks otsuse tegemise protsessi seoses potensiaalsete muutustega. Probleem antud meetodiga on, et simulatsioonimudeli loomine ja sobitamine on komplitseeritud ülesanne, mis vajab aega ja spetsialiseerunud tehnilisi teadmisi. Lisaks loovad analüütikud tavaliselt simulatsioonimudeleid, viies läbi intervjuusid, vaatlusi ja testimisi. Kõik need tehnikad on väga altid eelarvamustele, mis tähendab, et manuaalselt loodud mudelite täpsus on suhteliselt ekslik. Kõik see valmistab pettumust äriprotsessi simulatsiooni kasutusele võtmisel, mis teeb ettevõtetele antud tehnika kasutamise keeruliseks. Käesolev doktoritöö pakub välja uusi tehnikaid loomaks äriprotsessi simulatsioonimudeleid, mis kasutavad andmeid ettevõtete infosüsteemidest samaaegselt neuronvõrkude ja protsessikaeve algoritmidega. Antud doktoritöö eesmärk on luua täpsemat automaatset simulatsioonitehnikat, mis vajab vähem inimese sekkumist, lahendamaks puuduseid hetkel kasutuselolevast protsessi simulatsioonimootori lähenemisest. Me ühendame käesolevas doktoritöös välja toodud tehnikad kahes avatud lähtekoodiga tööriistas. Esimene tööriist, Simod, suudab täisautomaatselt avastada ja peenhäälestada simulatsioonimudeleid läbi kaeveprotsessi tehnikate. Välja toodud meetodil on siiski puudused, mis puudutavad iga tegevuse ajaennustust. Vastuseks on teine tööriist, DeepSimulator, mis ühildab avastamistehnikad, baseerudes kaeveprotsessile koos generatiivsete mudelitega, mis põhinevad süvaõppel. Hinnangu tulemused sellise hübriidlähenemise viisil viivad simulatsioonideni, mis peegeldavad lähemalt täheldatud protsessi dünaamikat kui meetodid, mis põhinevad paljalt kaeveprotsessil või süvaõppel.

Robotic Process Automation (RPA) is a technology to automate repetitive tasks. Using an RPA tool, a user can develop a software bot that can execute a sequence of interactive steps with one or more software applications. For example, we can use an RPA tool to develop a software bot that opens a spreadsheet, then opens a Web form, and then copy-pastes data from one of the cells in the spreadsheet into one of the fields in the Web form. While RPA tools allow users to automate many types of tasks, they do not help users to determine which tasks are candidates for automation. The current practice for identifying routines that are suitable for RPA automation is by means of interviews and observation of workers conducting their daily work. This manual approach is time-consuming and is not cost-efficient, especially in organizations that have a very large number of routines. In this thesis, we propose a set of techniques, which we call Robotic Process Mining (RPM) techniques, to discover automatable routines from recordings of interactions between one or more workers and one or more software applications. The goal of RPM techniques is to help business analysts to draw a systematic inventory of the tasks that can be automated in an organization. In addition, these techniques can produce scripts (programs) to automate each automatable task. The techniques proposed in this thesis have been consolidated in an open-source tool called Robidium. Robidium starts by recording one or more work sessions of several hours from one or more workers in an organization. Robidium then analyzes these recordings to discover sequences of steps that are frequently repeated (routines). Each routine is analyzed to determine if it can be automated using a software bot. When possible, Robidium produces a software bot capable of performing each routine automatically.

https://www.ester.ee/record=b5476326

Countries
Australia, Germany, Estonia
Related Organizations
Keywords

dissertations, dissertatsioonid, ETD, automatiseerimine, äriprotsessid, simulation, simulatsioon, 004, väitekirjad, Task mining, business processes, Process mining, automatic data processing, Robotic process automation, juhtimine

  • BIP!
    Impact byBIP!
    selected citations
    These citations are derived from selected sources.
    This is an alternative to the "Influence" indicator, which also reflects the overall/total impact of an article in the research community at large, based on the underlying citation network (diachronically).
    0
    popularity
    This indicator reflects the "current" impact/attention (the "hype") of an article in the research community at large, based on the underlying citation network.
    Average
    influence
    This indicator reflects the overall/total impact of an article in the research community at large, based on the underlying citation network (diachronically).
    Average
    impulse
    This indicator reflects the initial momentum of an article directly after its publication, based on the underlying citation network.
    Average
Powered by OpenAIRE graph
Found an issue? Give us feedback
selected citations
These citations are derived from selected sources.
This is an alternative to the "Influence" indicator, which also reflects the overall/total impact of an article in the research community at large, based on the underlying citation network (diachronically).
BIP!Citations provided by BIP!
popularity
This indicator reflects the "current" impact/attention (the "hype") of an article in the research community at large, based on the underlying citation network.
BIP!Popularity provided by BIP!
influence
This indicator reflects the overall/total impact of an article in the research community at large, based on the underlying citation network (diachronically).
BIP!Influence provided by BIP!
impulse
This indicator reflects the initial momentum of an article directly after its publication, based on the underlying citation network.
BIP!Impulse provided by BIP!
0
Average
Average
Average
Green