slice, mine, dice: complexity-aware automated discovery of business process models
DESCRIPTION
Presentation at the 11th International Conference on Business Process Management (BPM'2013), Beijing, China, 27 August 2013. Winner of the best paper award.TRANSCRIPT
![Page 1: Slice, Mine, Dice: Complexity-Aware Automated Discovery of Business Process Models](https://reader036.vdocuments.site/reader036/viewer/2022081602/54844d2cb4af9ff46c8b4667/html5/thumbnails/1.jpg)
1
Slice, Mine, Dice: Complexity-Aware Automated Discovery of Business Process Models
Chathura C. Ekanayake (QUT), Marlon Dumas (UT),Luciano García-Bañuelos (UT), Marcello La Rosa (QUT)
![Page 2: Slice, Mine, Dice: Complexity-Aware Automated Discovery of Business Process Models](https://reader036.vdocuments.site/reader036/viewer/2022081602/54844d2cb4af9ff46c8b4667/html5/thumbnails/2.jpg)
Useless!
Event log
Discovering “all-in-one” process models…
Process discovery algorithms
2
![Page 3: Slice, Mine, Dice: Complexity-Aware Automated Discovery of Business Process Models](https://reader036.vdocuments.site/reader036/viewer/2022081602/54844d2cb4af9ff46c8b4667/html5/thumbnails/3.jpg)
a b a b c d e c a f g h
a b a b k d e c h f g h
b c p q p r a k q r s
b c p h p r a k q r s
a x p h y z t t u
Trace clustering
a b a b c d e c a f g h
a b a b k d e c h f g h
b c p q p r a k q r s
b c p h p r a k q r s
a x p h y z t t u
Cluster 1
Cluster 2
Noise
Event log
Process variant 1
Process variant 2
Trace clustering
high complexity & redundancy
3
![Page 4: Slice, Mine, Dice: Complexity-Aware Automated Discovery of Business Process Models](https://reader036.vdocuments.site/reader036/viewer/2022081602/54844d2cb4af9ff46c8b4667/html5/thumbnails/4.jpg)
Process discovery
Trace clustering
Event log
Complexity thresholde.g. Size ≤ 30
Process model
Slice, Mine and Dice (SMD)
>?
4
1) Slice
2) Mine
3) Dice
![Page 5: Slice, Mine, Dice: Complexity-Aware Automated Discovery of Business Process Models](https://reader036.vdocuments.site/reader036/viewer/2022081602/54844d2cb4af9ff46c8b4667/html5/thumbnails/5.jpg)
Process discovery
Trace clustering
Event log
Complexity thresholde.g. Size ≤ 30
Slice, Mine and Dice (SMD)
>?Process models
5
![Page 6: Slice, Mine, Dice: Complexity-Aware Automated Discovery of Business Process Models](https://reader036.vdocuments.site/reader036/viewer/2022081602/54844d2cb4af9ff46c8b4667/html5/thumbnails/6.jpg)
Process discovery
Trace clustering
Event log
Complexity thresholde.g. Size ≤ 30
Slice, Mine and Dice (SMD)
>?Process models
6
![Page 7: Slice, Mine, Dice: Complexity-Aware Automated Discovery of Business Process Models](https://reader036.vdocuments.site/reader036/viewer/2022081602/54844d2cb4af9ff46c8b4667/html5/thumbnails/7.jpg)
Process model M3
Process model M2
A closer look…
7
![Page 8: Slice, Mine, Dice: Complexity-Aware Automated Discovery of Business Process Models](https://reader036.vdocuments.site/reader036/viewer/2022081602/54844d2cb4af9ff46c8b4667/html5/thumbnails/8.jpg)
F10
F11 F13
F14F12
M2 RPST of M2
F20
F21 F22
F24F23
F25
RPST of M3
Refined Process Structure Tree (RPST)
J. Vanhatalo, H. Volzer, J. Koehler: The Refined Process Structure Tree. Data Knowl. Eng., 2009
M3
8
![Page 9: Slice, Mine, Dice: Complexity-Aware Automated Discovery of Business Process Models](https://reader036.vdocuments.site/reader036/viewer/2022081602/54844d2cb4af9ff46c8b4667/html5/thumbnails/9.jpg)
M2M3
F14 F25
M. Dumas, L. García-Bañuelos, M. La Rosa, and R. Uba. Fast Detection of Exact Clones in Business Process Model Repositories. Information Systems, 2013
RPSDAG
F10
F11 F13
F14F12
RPST of M2
F20
F21 F22
F24F23
F25
RPST of M3
9
![Page 10: Slice, Mine, Dice: Complexity-Aware Automated Discovery of Business Process Models](https://reader036.vdocuments.site/reader036/viewer/2022081602/54844d2cb4af9ff46c8b4667/html5/thumbnails/10.jpg)
F14
M2
M3
S3
+
S3
+
Extracting exact clones
S3
Exact clones:• SESE• Non-trivial• Identical
10
![Page 11: Slice, Mine, Dice: Complexity-Aware Automated Discovery of Business Process Models](https://reader036.vdocuments.site/reader036/viewer/2022081602/54844d2cb4af9ff46c8b4667/html5/thumbnails/11.jpg)
F12
F22M3
+
S3
+S3
?
?
Extracting approximate clones
C. C. Ekanayake, M. Dumas, L. García-Bañuelos, M. La Rosa, and A. H. M. ter Hofstede. Approximate Clone Detection in Repositories of Business Process Models, BPM 2012
Appr. clones:• SESE• Non-trivial• Similar• Unrelated
11
+
+
M2
![Page 12: Slice, Mine, Dice: Complexity-Aware Automated Discovery of Business Process Models](https://reader036.vdocuments.site/reader036/viewer/2022081602/54844d2cb4af9ff46c8b4667/html5/thumbnails/12.jpg)
Merging algorithm
Fragment F12 of model M2
Fragment F22 of model M3
Configurable gateway
Configurable label
M. La Rosa, M. Dumas, R. Uba, and R. M. Dijkman. Business Process Model Merging: An Approach to Business Process Consolidation. ACM TOSEM, 2013.
Merging approximate clones
S4
12
![Page 13: Slice, Mine, Dice: Complexity-Aware Automated Discovery of Business Process Models](https://reader036.vdocuments.site/reader036/viewer/2022081602/54844d2cb4af9ff46c8b4667/html5/thumbnails/13.jpg)
Consolidated process model collection
S4S3
+
The result…
13
+
+
+
![Page 14: Slice, Mine, Dice: Complexity-Aware Automated Discovery of Business Process Models](https://reader036.vdocuments.site/reader036/viewer/2022081602/54844d2cb4af9ff46c8b4667/html5/thumbnails/14.jpg)
Trace clustering• M. Song, C.W. Gunther, and W.M.P. van der Aalst, Improving Process Mining with
Trace Clustering, J. Korean Inst. of Industrial Engineers 34(4), 2008• R.P.J.C. Bose, W.M.P. van der Aalst, Trace Clustering Based on Conserved Patterns:
Towards Achieving Better Process Models, BPM 2009 Workshops• A.K.A. de Medeiros, A. Guzzo, G. Greco, W.M.P. van der Aalst, A.J.M.M. Weijters,
B.F. van Dongen, D. Saccà. Process Mining Based on Clustering: A Quest for Precision, BPM Workshops 2007
Discovery• A.J.M.M. Weijters, J.T.S. Ribeiro. Flexible Heuristics Miner (FHM), CIDM, 2011.
Evaluation setup
Log Traces Events Eventclasses
Duplication ratio
Motor 4,293 33,202 292 114
Commercial 4,852 54,134 81 668
BPI 2012 5,312 91,949 36 2,554
14
![Page 15: Slice, Mine, Dice: Complexity-Aware Automated Discovery of Business Process Models](https://reader036.vdocuments.site/reader036/viewer/2022081602/54844d2cb4af9ff46c8b4667/html5/thumbnails/15.jpg)
Evaluation – repository size and models number S: Song et al. B: Bose et al. M: Medeiros et al.
• up to 64% reduction in repository size• up to 66% reduction in # of top level process models• up to 120 sub-processes extracted
Motor Comm BPI Motor Comm BPI
14% 22%
66%
15
64%
![Page 16: Slice, Mine, Dice: Complexity-Aware Automated Discovery of Business Process Models](https://reader036.vdocuments.site/reader036/viewer/2022081602/54844d2cb4af9ff46c8b4667/html5/thumbnails/16.jpg)
Evaluation – individual model complexityLog Method Size CFC ACD Density CNC
Avg Min Max Savings (%)
Motor
S 22.75 4 37 22.8 12.07 2.71 0.07 1.26S + SMD 17.57 4 37 10.07 2.34 0.11 1.21B 20.01 4 37 9.8 9.97 2.51 0.08 1.2B + SMD 18.04 4 37 10.05 2.38 0.11 1.2M 15.73 3 49 -1.1 7.36 2.14 0.11 1.12M + SMD 15.9 4 49 8.34 2.12 0.12 1.14
Commercial
S 24.07 6 34 22.4 13.65 2.96 0.06 1.32S + SMD 18.67 2 34 11.34 2.49 0.1 1.24B 21.11 2 34 20.3 11.04 2.65 0.07 1.23B+ SMD 16.82 2 34 9.73 2.29 0.12 1.18M 18.86 2 40 11.1 10.18 2.47 0.09 1.22M + SMD 16.76 2 34 9.71 2.38 0.11 1.21
BPI
S 47.32 15 56 29.7 20.77 2.34 0.03 1.24S + SMD 33.27 4 56 20.18 2.41 0.07 1.28B 46.54 13 56 30.6 20.48 2.35 0.03 1.23B + SMD 32.3 4 56 19.29 2.33 0.07 1.27M 46.48 21 61 18.9 21.16 2.34 0.03 1.24M + SMD 37.71 7 56 25.29 2.38 0.04 1.3
16
![Page 17: Slice, Mine, Dice: Complexity-Aware Automated Discovery of Business Process Models](https://reader036.vdocuments.site/reader036/viewer/2022081602/54844d2cb4af9ff46c8b4667/html5/thumbnails/17.jpg)
+ Seeks to discover process models that meet user-specified complexity thresholds
+ Reduces repository size and top level models number compared to trace clustering techniques
+ Preserves fitness, appropriateness & generalization of models discovered from trace clusters
+ Little impact on structural complexity of process models
- Performance overhead
The SMD technique
17
![Page 18: Slice, Mine, Dice: Complexity-Aware Automated Discovery of Business Process Models](https://reader036.vdocuments.site/reader036/viewer/2022081602/54844d2cb4af9ff46c8b4667/html5/thumbnails/18.jpg)
Optimizations• Parallelization of divisive trace clustering, discovery and
GED computation• Incremental divisive trace clustering• Sub-polygons and sub-bonds extraction
Complexity thresholds tuning
In the future…
18
![Page 19: Slice, Mine, Dice: Complexity-Aware Automated Discovery of Business Process Models](https://reader036.vdocuments.site/reader036/viewer/2022081602/54844d2cb4af9ff46c8b4667/html5/thumbnails/19.jpg)
Slice, Mine, Dice: Complexity-Aware Automated Discovery of Business Process Models