A Methodology to Predict the Performance of Distributed Simulation Systems
Bpm10slides
1. A fresh look at Precision in Process Conformance Jorge Muñoz-GamaJosep Carmona UniversitatPolitècnica de Catalunya (Barcelona, Spain)
2. Outline 15 Sep 2010 Precision in Process Conformance 2 Process Mining and Process Conformance Motivation Approach General Approach Implementation Results Extensions Future work Conclusions
3. Process Mining 15 Sep 2010 Precision in Process Conformance 3 * www.processmining.org
4. Conformance Dimensions 15 Sep 2010 Precision in Process Conformance 4 How much of the observed behavior is captured by the model Models with minimal behavior to represent accurately the log Overly precise models which overfit the log Minimal structure which clearly reflect the behavior
5. Outline 15 Sep 2010 Precision in Process Conformance 5 Process Mining and Process Conformance Motivation Approach General Approach Implementation Results Extensions Future work Conclusions
6. Related Work 15 Sep 2010 Precision in Process Conformance 6 Precision in the literature Most related work Rozinat et al. Information System 33 (2008) Metric for Precision in Petri Nets Computation of Follows and Precedes relations (Always, Never, Sometimes) of Model and Log. Measurement based on discrepancies in Sometimes relations Model relations require a model space state exploration Coverability Graph
7. Motivation 15 Sep 2010 Precision in Process Conformance 7 Goals and Requirements Precision Dimension Petri Nets Avoid the complete state space exploration Effort needed to obtain an accurate model Fine-level precision Locate the precision inconsistencies
8. Process Conformance and Refinement 15 Sep 2010 Precision in Process Conformance 8 Locate the inconsistencies Petri Net Conformance (Precision) B A D C MDT ETC Precision Metric A B D A C D Measure the inconsistencies Event Log
9. Outline 15 Sep 2010 Precision in Process Conformance 9 Process Mining and Process Conformance Motivation Approach General Approach Implementation Results Extensions Future work Conclusions
10. General Idea: Escaping Edges 15 Sep 2010 Precision in Process Conformance 10 Model Behavior Escaping Edges Log Behavior Model Behavior
11. Conformance Route Map 15 Sep 2010 Precision in Process Conformance 11 Petri Net B A D C MDT Model States Traversal Metric Log States A B D A C D Event Log
12. Log and Model States 15 Sep 2010 Precision in Process Conformance 12 Log Incorporate state information in the log (Aalst et al. Software and Systems Modeling, 2009) Past, Unlimited and Sequence Model Markings of the Petri Net
13. Model States and Mapping 15 Sep 2010 Precision in Process Conformance 13 Not all the reachable markings (could be infinite) Only Markings with a Log State mapped on Log and Model States Mapping i.e., reached marking after replay state prefix p2 p3 p4 p1 p1 p2 p3 p4 p5 B A B E s2 s1 s3 s4 0 1 0 0 1 … 0 1 0 0 n E A C p5 D Markings not explored p1 p4 p3 p2
14. Log-guided Traversal 15 Sep 2010 Precision in Process Conformance 14 Log-guided Traversal of Model Behavior Allowed Tasks : i.e., actions enabled in that moment Reflected Tasks : i.e., actions really executed (thus, annotated in the log) B C D <p2> p2 p3 p4 p1 A B E A C E B B B C p2 p3 p4 p1 E E A C A C A B E A C E D D
15. Traversal (2) 15 Sep 2010 Precision in Process Conformance 15 Escaping Edges : i.e., enabled actions not executed Precision discrepancies B C D B p1 p2 p3 p4 E A C B C D A B E A C E D
16. Precision Metric 15 Sep 2010 Precision in Process Conformance 16 Take into account the Escaping Edges Between 0 (imprecise) and 1 (precise) More frequent traces, more weight in the metric Independent of Structural dimension Global precision Localizability A H I Z A P Q Z H I A Z P Q
17. Minimal Disconformant Traces (MDT) 15 Sep 2010 Precision in Process Conformance 17 Localizability of precision inconsistencies i.e., Minimal traces indicating where the model starts to deviate from the log Algorithm to compute all MDT using Escaping Edges B D A C MDT A E A B E C D P Q Refined Petri Net
18. Outline 15 Sep 2010 Precision in Process Conformance 18 Process Mining and Process Conformance Motivation Approach General Approach Implementation Results Extensions Future work Conclusions
19. Implementation 15 Sep 2010 Precision in Process Conformance 19 ProM 6 Framework ETConformance Plug-In
20. Outline 15 Sep 2010 Precision in Process Conformance 20 Process Mining and Process Conformance Motivation Approach General Approach Implementation Results Extensions Future work Conclusions
30. Invisible Tasks (2) 15 Sep 2010 Precision in Process Conformance 25 Invisible Coverability Graph Solutions Union of Enabled Lazy Invisibles * One path only Shortest Invisible Path * A,B A B <1, 0, 0> Inv2 Inv1 <1, ω, 0> <0, 0, 1> C C Inv3 A,D D X <0, ω, 1> A,C X X *Rozinat et al. Information System 33 (2008)
31. Duplicate Tasks 15 Sep 2010 Precision in Process Conformance 26 (Several Transitions associated with the same event) Which Task? B ? B ? INDETERMINISM Solutions e.g. Look-ahead B C A B D ... A B C ...
32.
33. Variant: Non fitting models 15 Sep 2010 Precision in Process Conformance 28 Symmetric to the Escaping Edges (Ee) Log Escaping Edges (LEe): The points where the log deviates from the model Fitness instead of Precision Model Behavior Escaping Edges Log Behavior Log Escaping Edges Model Behavior
34. Outline 15 Sep 2010 Precision in Process Conformance 29 Process Mining and Process Conformance Motivation Approach General Approach Implementation Results Extensions Future work Conclusions
35. Future Work: Refinement 15 Sep 2010 Precision in Process Conformance 30 B A D C MDT A E A B E B H J G Refined Petri Net Event Log B A E A B E D A Petri Net C
36. Concurrencies in the model but not in the log Break the model concurrency with a restriction, e.g. a place Structural Concurrency Best effort overapproximation for general Petri Nets Exact for live and bounded Free Choice systems Polynomial Algorithm Kovalyov and Esparza , Proc. Intl. Workshop on Discrete Event Sytems, 1996 Future Work: Breaking Concurrencies 15 Sep 2010 Precision in Process Conformance 31 B D A A B C D C
37. Supervisory Control 15 Sep 2010 Precision in Process Conformance 32 Refined Model MDT Abstraction MDT Supervisor Model Supervisory Control in Process Mining Santos et al. Supervisory Control Service (2010)
38. Conclusions 15 Sep 2010 Precision in Process Conformance 33 New technique for precision between Petri nets and Log. Avoids complete models state space exploration. Metric based on the effort needed to obtain a precise model. MDT, indicating the points where the model starts to deviates from the log. Approach implemented as plug-in of ProM 6.
39. Thank You 15 Sep 2010 Precision in Process Conformance 34 Thank You for Your Attention