Presentation is loading. Please wait.

Presentation is loading. Please wait.

A Framework for Recommending Resource Allocation Based on Process Mining MICHAEL ARIAS ERIC ROJAS JORGE MUNOZ-GAMA MARCOS SEPÚLVEDA.

Similar presentations


Presentation on theme: "A Framework for Recommending Resource Allocation Based on Process Mining MICHAEL ARIAS ERIC ROJAS JORGE MUNOZ-GAMA MARCOS SEPÚLVEDA."— Presentation transcript:

1 A Framework for Recommending Resource Allocation Based on Process Mining MICHAEL ARIAS ERIC ROJAS JORGE MUNOZ-GAMA MARCOS SEPÚLVEDA

2 Outline A Framework for Recommending Resource Allocation Based on Process Mining 2 Introduction Approach ◦ General approach ◦ Resource allocation request ◦ Resource process cube and allocation metrics ◦ Implementation Experimental evaluation Future work Conclusions

3 Outline A Framework for Recommending Resource Allocation Based on Process Mining 3 Introduction Approach ◦ General approach ◦ Resource allocation request ◦ Resource process cube and allocation metrics ◦ Implementation Experimental evaluation Future work Conclusions

4 Motivation A Framework for Recommending Resource Allocation Based on Process Mining 4 Resources Business processe s Cost reduction Improve performanc e

5 Motivation A Framework for Recommending Resource Allocation Based on Process Mining 5

6 Related work Human resource allocation: an important issue in Business Process Management A Framework for Recommending Resource Allocation Based on Process Mining 6 [1] Russell et al. 2005 [5] Liu et al. 2008 [3] Huang et al. 2011 [4] Liu et al. 2012 [2] Rinderle-Ma and van der Aalst. 2007 [6] Huang et al. 2011 [7] Koschmider et al. 2011 [8] Cabanillas et al. 2013

7 Related work A Framework for Recommending Resource Allocation Based on Process Mining 7 Activity profile ✓✓ Resource profile ✓✓✓✓✓ Performance & quality ✓ Resource meta-model ✓✓✓✓ History ✓✓✓✓✓✓✓✓✓ Process mining tool ✓✓✓ Allocation at sub-process level ✓

8 Related work A Framework for Recommending Resource Allocation Based on Process Mining 8 [9] Zhao, W., Zhao, X.: Process mining from the organizational perspective, 2014

9 Related work A Framework for Recommending Resource Allocation Based on Process Mining 9 Activity profile ✓✓ Resource profile ✓✓✓✓✓ Performance & quality ✓ Resource meta-model ✓✓✓✓ History ✓✓✓✓✓✓✓✓✓ Process mining tool ✓✓✓ Allocation at sub-process level ✓

10 General idea A Framework for Recommending Resource Allocation Based on Process Mining 10 Resource profile Performance & Quality History Process mining tool Activity profile Allocation at sub-process level Allocate the most appropriate resource Framework for recommending resource allocation Resource meta-model

11 Outline A Framework for Recommending Resource Allocation Based on Process Mining 11 Introduction Approach ◦ General approach ◦ Resource allocation request ◦ Resource process cube and allocation metrics ◦ Implementation Experimental evaluation Future work Conclusions

12 Proposed framework A Framework for Recommending Resource Allocation Based on Process Mining 12 Character istics Informat ion Weights 1 1

13 Proposed framework A Framework for Recommending Resource Allocation Based on Process Mining 13 Character istics Informat ion Weights 2 2

14 Proposed framework A Framework for Recommending Resource Allocation Based on Process Mining 14 Character istics Informat ion Weights 3 3

15 Proposed framework A Framework for Recommending Resource Allocation Based on Process Mining 15 Character istics Informat ion Weights 4 4

16 Proposed framework A Framework for Recommending Resource Allocation Based on Process Mining 16 Character istics Informat ion Weights 5 5

17 Proposed framework A Framework for Recommending Resource Allocation Based on Process Mining 17 Character istics Informat ion Weights 1 1 2 2 3 3 4 4 5 5

18 Help-Desk process A Framework for Recommending Resource Allocation Based on Process Mining 18 Printer Support Server Support Database Support Desktop Support Help Desk First-contact level Ring…!!! I have a issue

19 Resource allocation request A Framework for Recommending Resource Allocation Based on Process Mining 19 Character istics Informat ion Weights 1 1

20 Resource allocation request A Framework for Recommending Resource Allocation Based on Process Mining 20 Characterize the request Characterization c = (f 1, …, f n ) c = sub-process, typology Characteristics In the Help-Desk: First contact level 1 Expert level 2 Printer-support Server-support Typologies

21 Resource allocation request A Framework for Recommending Resource Allocation Based on Process Mining 21 Character istics Informat ion Weights 2 2 3 3 1 1

22 Resource allocation request A Framework for Recommending Resource Allocation Based on Process Mining 22 Character istics Informat ion Weights 2 2 3 3 1 1

23 Resource allocation request A Framework for Recommending Resource Allocation Based on Process Mining 23 Characterize the request Proposed dimensions Expertise matrices Information Frequency Performance Quality Cost Workload Expertise Dimension Competencies Dimensions (history) Resource process cube ( Q ) Skills Knowledge E c [1 : n] E r [1 : n] D Example E c 1 = [2,2] E r 1 = [0,1] Required Expertise Resource Expertise C 1 = (sub-process 1, printers) Require d: Resource 1:

24 Resource process cube (Q) A Framework for Recommending Resource Allocation Based on Process Mining 24 R1 R2 Rn … Characteristic s Resources Dimensions Frequency Performance Quality Cost Workload Sub-process, typology

25 Resource allocation metrics A Framework for Recommending Resource Allocation Based on Process Mining 25 Performance_Met ric (r,c) = [ ][c][ p ].max - [r][c][ p ].avg Q Q [ ][c][ p ].max - [ ][c][ p ].min Q Q Prioritize the fastest resources

26 Resource allocation metrics A Framework for Recommending Resource Allocation Based on Process Mining 26 Quality_Metric (r,c) = [r][c][ q ].avg - [ ][c][ q ].min Q Q [ ][c][ q ].max - [ ][c][ q ].min Q Q Prioritize the best evaluated resources

27 Resource allocation metrics A Framework for Recommending Resource Allocation Based on Process Mining 27 Cost_Metric (r,c) = [ ][c][ co ].max - [r][c][ co ].avg Q Q [ ][c][ co ].max - [ ][c][ co ].min Q Q Prioritize the cheapest resources

28 Resource allocation metrics A Framework for Recommending Resource Allocation Based on Process Mining 28 Workload_Metric (r,c) = [r][ ][ w ].top - [r][ ][ w ].total Q Q [r][ ][ w ].top - [r][ ][ w ].bottom Q Q Prioritize the most idle resources

29 Resource allocation metrics A Framework for Recommending Resource Allocation Based on Process Mining 29 Frequency_Metri c (r,c) = logarithm ( [r][c][ f ].total) +1 Q logarithm ( [ ][c][ f ].total) +1 Q Prioritize the most experienced resources

30 Resource allocation metrics A Framework for Recommending Resource Allocation Based on Process Mining 30 1-1- 1n1n √ ∑ n (under(i)) 2 i =1 underQualification_ Metric = under(i ) = c [i]- r [i] c [i] - i EE E if c [i] >= r [i] EE 0 otherwi se Identify how below the resource is from the desired expertise level

31 Resource allocation metrics A Framework for Recommending Resource Allocation Based on Process Mining 31 1-1- 1n1n √ ∑ n (over(i)) 2 i =1 overQualification_ Metric = over(i) = r [i]- c [i] - c [i] i E E E if r [i] >= c [i] E E 0 otherwi se Identify how above the resource is from the desired expertise level

32 Resource allocation request A Framework for Recommending Resource Allocation Based on Process Mining 32 Character istics Informat ion Weights 2 2 3 3 1 1

33 Resource allocation request A Framework for Recommending Resource Allocation Based on Process Mining 33 Characterize the request Describe the importance of each dimension Weights Large size company: F:010 P:050 Q:010 C:10 0 U:015 O:000 Small size company: F:025 P:015 Q:100 C:03 0 U:075 O:065

34 Proposed framework A Framework for Recommending Resource Allocation Based on Process Mining 34 Character istics Informat ion Weights 2 2 3 3 1 1 4 4 5 5

35 Implementation with BPA [10] A Framework for Recommending Resource Allocation Based on Process Mining 35 Frequency Data ItemLocal Score R60.601 R40.593 R50.554 R10.525 …… Performance Data ItemLocal Score R30.851 R20.833 R10.832 R50.775 …… Quality Data ItemLocal Score R90.913 R70.864 R80.808 R20.723 …… … …… …… Position 1 2 3 4 … … …… …… … …… …… Overall top Data itemScore R10.795 R20.788 R140.784 [10] Akbarinia, R., Pacitti, E., Valduriez, P.: Best position algorithms for ecient top-k query processing, 2011

36 Outline A Framework for Recommending Resource Allocation Based on Process Mining 36 Introduction Approach ◦ General approach ◦ Resource allocation request ◦ Resource process cube and allocation metrics ◦ Implementation Experimental evaluation Future work Conclusions

37 Experimental evaluation A Framework for Recommending Resource Allocation Based on Process Mining 37

38 A Framework for Recommending Resource Allocation Based on Process Mining 38 Top 3-queries Single metric Experiment 1 Experiments results

39 A Framework for Recommending Resource Allocation Based on Process Mining 39 Top 3-queries 3 help desk companies Experiment 2 Experiments results   

40 A Framework for Recommending Resource Allocation Based on Process Mining 40 Top 3-queries 3 help desk companies Log size Amount of resources Experiment 3 Experiments results

41 A Framework for Recommending Resource Allocation Based on Process Mining 41 Performance analysis ( a ) Number of cases (thousands) ( b ) Number of resources Experiments results

42 Outline A Framework for Recommending Resource Allocation Based on Process Mining 42 Introduction Approach ◦ General approach ◦ Resource allocation request ◦ Resource process cube and allocation metrics ◦ Implementation Experimental evaluation Future work Conclusions

43 Future work A Framework for Recommending Resource Allocation Based on Process Mining 43 Case studies using real data Incorporate new dimensions Explore potential application domains Combine our approach with existing works

44 Outline A Framework for Recommending Resource Allocation Based on Process Mining 44 Introduction Approach ◦ General approach ◦ Resource allocation request ◦ Resource process cube and allocation metrics ◦ Implementation Experimental evaluation Future work Conclusions

45 A Framework for Recommending Resource Allocation Based on Process Mining 45 Consider different perspectives New allocation technique Multi – factor criteria Sub-process level Fine-grained, generic, & extensible Experimental evaluation Synthetic data BPA allows obtain a final ranking

46 THANK YOU FOR YOUR ATTENTION

47 A Framework for Recommending Resource Allocation Based on Process Mining MICHAEL ARIAS ERIC ROJAS JORGE MUNOZ-GAMA MARCOS SEPÚLVEDA

48 ANNEX 1 NEW DIMENSIONS

49 New dimensions A Framework for Recommending Resource Allocation Based on Process Mining 49 ITAT by priority (Performance) First Time Fix (Quality) # of incidents by priority (Compliance) Resolved within SLA by priority (Value) % Reduction incidents (Value) # of incidents resolved by… (Quality) # of proactive problems (Compliance) Average time to provide workaround by priority (Efficiency) Incident management dimension Problem management dimension

50 ANNEX 2 BEST POSITION ALGORITHM

51 Best Position Algorithm (BPA) A Framework for Recommending Resource Allocation Based on Process Mining 51

52 BPA (iteration 1) A Framework for Recommending Resource Allocation Based on Process Mining 52 - Sorted Access to Pos 1: {d1, d2, d3} - Random Access to obtain local score y position. P1 1 = {1, 4, 9} and Bp 1 = 1 P1 2 = {1, 6, 8} and Bp 2 = 1 P1 3 = {1, 5, 8} y and Bp 3 = 1 Best position score: λ = f (S 1 (1) + S 2 (1) + S 3 (1)) = 30 + 28 + 30 = 88

53 BPA (iteration 1) A Framework for Recommending Resource Allocation Based on Process Mining 53 L 1 P 1 = {1, 4, 9} and BP 1 = 1 L 2 P 2 = {1, 6, 8} and BP 2 = 1 L 3 P 3 = {1, 5, 8} and BP 3 = 1 λ = f (S 1 (1), S 2 (1), S 3 (1)) = 30 + 28 +30 = 88. λ be the Threshold Y = {d1, d2, d3}, Y visited elements Z = {65, 63, 70}, Z score global Best position score = 88 > overall score of the highest data items {d1, d2, d3} (65,63,70) → continue

54 BPA (iteration 2) A Framework for Recommending Resource Allocation Based on Process Mining 54 - Sorted Access to Pos 2: {d4, d5, d6} - Random Access to obtain local score and position. P 21 = {2,7,8} and Bp 21 = 2 P 22 = {2,4,9} and Bp 22 = 2 P 23 = {2,4,6} and Bp 23 = 2 Best position score: λ = f (S 1 (2) + S 2 (2) + S 3 (2)) = 28 + 27 + 29 = 84

55 BPA (iteration 3) A Framework for Recommending Resource Allocation Based on Process Mining 55 - Sorted Access to Pos 3: {d9, d7, d8} - Random Access to obtain local score and position. P3 1 = {3,5,6} and Bp 31 = 9 P3 2 = {3,5,7} and Bp 32 = 9 P3 3 = {3,9,10} and Bp 33 = 6 Best position score: λ = f (S 1 (9) + S 2 (9) + S 3 (6)) = 11 + 13 + 19 = 43

56 BPA (iteration 3) A Framework for Recommending Resource Allocation Based on Process Mining 56 P 31 = {3, 5, 6} Bp 1 = 9, P 32 = {3, 5, 7} Bp 2 = 9, P 33 = {3, 9, 10} Bp 3 = 6 PL 1 = PIt 1 L 1 U PIt 2 L 1 U PIt 3 L 1 = {1,2,3,4,5,6,7,8,9} PL 2 = PIt 1 L 2 U PIt 2 L 2 U PIt 3 L 2 = {1,2,3,4,5,6,7,8,9} PL 3 = PIt 1 L 3 U PIt 2 L 3 U PIt 3 L 3 = {1,2,3,4,5,6,8,9,10} λ = f (S1(9), S2(9), S3(6)) = 11 + 13 + 19 = 43 Last Global Score: {d3, d4, d6} = {70, 66, 70} Current Global Score : {d9, d7, d8} = {62, 61, 71} Update Y. Y = {d3, d6, d8}, = {70, 70, 71} The values in Y are greater than λ= 43, the stop condition finish in position 3.

57 BPA (iteration 3) A Framework for Recommending Resource Allocation Based on Process Mining 57


Download ppt "A Framework for Recommending Resource Allocation Based on Process Mining MICHAEL ARIAS ERIC ROJAS JORGE MUNOZ-GAMA MARCOS SEPÚLVEDA."

Similar presentations


Ads by Google