| Presentation | 2023-03-17 Analyzing Business Processes by Automatically Detecting KPI Thresholds Based on Trace Variants Taro Takei, Hiroki Horita, |
|---|---|
| PDF Download Page |  PDF download Page Link |
| Abstract(in Japanese) | (See Japanese page) |
| Abstract(in English) | One method for analyzing complex business processes is to filter event logs by KPI thresholds to extract only specific processes; processes filtered by KPI thresholds become simpler and easier to understand than the original processes. However, existing research has proposed an automatic KPI threshold detection method, which reduces the time and effort required to find a threshold value. In the existing method, a process model is first generated by dividing the log by an arbitrary number of divisions $k$. Then, by repeatedly aggregating the most similar models, the domain conditions of the final process model are extracted as KPI thresholds. However, the existing method requires trial-and-error to determine the value of $k$, which is time-consuming because the threshold detected and the process model change depending on the value of $k$, the number of log partitions. In this paper, we propose a method to automatically detect KPI thresholds by dividing logs based on trace variants. By dividing the log by variants, we avoid mixing dissimilar processes and further reduce the threshold detection time. Experimental results show that the process models detected by the proposed method are simpler and the threshold detection time is significantly reduced. |
| Keyword(in Japanese) | (See Japanese page) |
| Keyword(in English) | business process analysis / process mining / process discovery / KPI / graph edit distance |
| Paper # | KBSE2022-66 |
| Date of Issue | 2023-03-09 (KBSE) |
| Conference Information | |
| Committee | KBSE |
|---|---|
| Conference Date | 2023/3/16(2days) |
| Place (in Japanese) | (See Japanese page) |
| Place (in English) | JMS ASTERPLAZA |
| Topics (in Japanese) | (See Japanese page) |
| Topics (in English) | |
| Chair | Takuya Saruwatari(NTT Data) |
| Vice Chair | Yoshinori Tanabe(Tsurumi Univ.) |
| Secretary | Yoshinori Tanabe(Osaka Inst. of Tech.) |
| Assistant | Yoshitaka Aoki(BIPROGY) / Hiroki Horita(Ibaraki Univ.) |
| Paper Information | |
| Registration To | Technical Committee on Knowledge-Based Software Engineering |
|---|---|
| Language | JPN |
| Title (in Japanese) | (See Japanese page) |
| Sub Title (in Japanese) | (See Japanese page) |
| Title (in English) | Analyzing Business Processes by Automatically Detecting KPI Thresholds Based on Trace Variants |
| Sub Title (in English) | |
| Keyword(1) | business process analysis |
| Keyword(2) | process mining |
| Keyword(3) | process discovery |
| Keyword(4) | KPI |
| Keyword(5) | graph edit distance |
| 1st Author's Name | Taro Takei |
| 1st Author's Affiliation | Ibaraki University(Ibaraki Univ.) |
| 2nd Author's Name | Hiroki Horita |
| 2nd Author's Affiliation | Ibaraki University(Ibaraki Univ.) |
| Date | 2023-03-17 |
| Paper # | KBSE2022-66 |
| Volume (vol) | vol.122 |
| Number (no) | KBSE-444 |
| Page | pp.pp.73-78(KBSE), |
| #Pages | 6 |
| Date of Issue | 2023-03-09 (KBSE) |