A Framework for Recommending
Resource Allocation Based
on Process Mining
MICHAEL ARIAS
ERIC ROJAS
JORGE MUNOZ-GAMA
M A R C O S S E P Ú LV E D A
Outline
Introduction
Approach
◦ General approach
◦ Resource allocation request
◦ Resource process cube and allocation metrics
◦ Implementation
Experimental evaluation
Future work
Conclusions
A Framework for Recommending Resource Allocation Based on Process Mining
2
Outline
Introduction
Approach
◦ General approach
◦ Resource allocation request
◦ Resource process cube and allocation metrics
◦ Implementation
Experimental evaluation
Future work
Conclusions
A Framework for Recommending Resource Allocation Based on Process Mining
3
Motivation
Cost
reduction
Improve
performance
Resources
Business
processes
A Framework for Recommending Resource Allocation Based on Process Mining
4
Motivation
A Framework for Recommending Resource Allocation Based on Process Mining
5
Related work
Human resource allocation: an important issue in
Management
Business Process
[6] Huang et al. 2011
[1] Russell et al. 2005
[7] Koschmider et al. 2011
[5] Liu et al. 2008
[2] Rinderle-Ma and van der Aalst. 2007
[8] Cabanillas et al. 2013
[3] Huang et al. 2011
[4] Liu et al. 2012
A Framework for Recommending Resource Allocation Based on Process Mining
6
Related work
✓
Activity profile
Resource profile
✓
✓
✓
✓
✓
✓
Performance & quality
Resource meta-model
✓
History
✓
Process mining tool
✓
✓
✓
✓
✓
✓
✓
✓
Allocation at sub-process level
A Framework for Recommending Resource Allocation Based on Process Mining
✓
✓
✓
✓
✓
✓
✓
✓
7
Related work
[9] Zhao, W., Zhao, X.: Process mining from the organizational perspective, 2014
A Framework for Recommending Resource Allocation Based on Process Mining
8
Related work
✓
Activity profile
Resource profile
✓
✓
✓
✓
✓
✓
Performance & quality
Resource meta-model
✓
History
✓
Process mining tool
✓
✓
✓
✓
✓
✓
✓
✓
Allocation at sub-process level
A Framework for Recommending Resource Allocation Based on Process Mining
✓
✓
✓
✓
✓
✓
✓
✓
9
General idea
Allocate the most appropriate resource
Activity profile
Resource profile
Performance & Quality
Resource meta-model
History
Process mining tool
Allocation at
process level
sub-
Framework for recommending
resource allocation
A Framework for Recommending Resource Allocation Based on Process Mining
10
Outline
Introduction
Approach
◦ General approach
◦ Resource allocation request
◦ Resource process cube and allocation metrics
◦ Implementation
Experimental evaluation
Future work
Conclusions
A Framework for Recommending Resource Allocation Based on Process Mining
11
Proposed framework
1
Humanresources
information
Characteri
stics
Informati
on
Weights
ResourceAllocationRequest
Event log
Resource
ProcessCube
BPA
Best PositionAlgorithm
RecommenderSystem
A Framework for Recommending Resource Allocation Based on Process Mining
12
Proposed framework
2
Humanresources
information
Characteri
stics
Informati
on
Weights
ResourceAllocationRequest
Event log
Resource
ProcessCube
BPA
Best PositionAlgorithm
RecommenderSystem
A Framework for Recommending Resource Allocation Based on Process Mining
13
Proposed framework
Humanresources
information
Characteri
stics
Informati
on
Weights
ResourceAllocationRequest
Event log
3
Resource
ProcessCube
BPA
Best PositionAlgorithm
RecommenderSystem
A Framework for Recommending Resource Allocation Based on Process Mining
14
Proposed framework
Humanresources
information
Characteri
stics
Informati
on
Weights
ResourceAllocationRequest
Event log
Resource
ProcessCube
4
BPA
Best PositionAlgorithm
RecommenderSystem
A Framework for Recommending Resource Allocation Based on Process Mining
15
Proposed framework
Humanresources
information
Characteri
stics
Informati
on
Weights
ResourceAllocationRequest
Event log
Resource
ProcessCube
5
BPA
Best PositionAlgorithm
RecommenderSystem
A Framework for Recommending Resource Allocation Based on Process Mining
16
Proposed framework
1
2
Humanresources
information
Characteri
stics
Informati
on
Weights
ResourceAllocationRequest
Event log
3
Resource
ProcessCube
4
5
BPA
Best PositionAlgorithm
RecommenderSystem
A Framework for Recommending Resource Allocation Based on Process Mining
17
Help-Desk process
Ring…!!!
I have a issue
Help Desk
Printer
Support
Server
Support
First-contact level
Database
Support
A Framework for Recommending Resource Allocation Based on Process Mining
Desktop
Support
18
Resource allocation request
1
Characteri
stics
Informati
on
Weights
ResourceAllocationRequest
A Framework for Recommending Resource Allocation Based on Process Mining
19
Resource allocation request
Characterize
the request
Characteristics
In the Help-Desk:
First contact level 1
Expert level 2
Characterization c = (f1, …, fn)
c = sub-process, typology
Typologies
Printer-support
Server-support
A Framework for Recommending Resource Allocation Based on Process Mining
20
Resource allocation request
1
2
Humanresources
information
Characteri
stics
Informati
on
Weights
ResourceAllocationRequest
Event log
3
Resource
ProcessCube
A Framework for Recommending Resource Allocation Based on Process Mining
21
Resource allocation request
1
2
Humanresources
information
Characteri
stics
Informati
on
Weights
ResourceAllocationRequest
Event log
3
Resource
ProcessCube
A Framework for Recommending Resource Allocation Based on Process Mining
22
Resource allocation request
Characterize
the request
Information
D Dimensions (history)
Expertise Dimension
Competencies
Frequency
Skills
Knowledge
Performance
Proposed dimensions
Expertise matrices
Quality
Cost
Workload
Required Expertise
Ec [1 : n]
Resource Expertise
Er [1 : n]
Example
C1 = (sub-process 1, printers)
Resource process cube (Q)
A Framework for Recommending Resource Allocation Based on Process Mining
Required:
Resource 1:
E c1 = [2,2]
Er1 = [0,1]
23
Resource process cube (Q)
Dimensions
Frequency
Performance
Quality
Cost
Workload
A Framework for Recommending Resource Allocation Based on Process Mining
24
Resource allocation metrics
Prioritize the fastest resources
[ ][c][ p ].max
QPerformance_Metri Q
].avg
Q[ ][c][ p ].max
Qc (r,c) =
[r][c][ p
[ ][c][ p
].min
A Framework for Recommending Resource Allocation Based on Process Mining
25
Resource allocation metrics
Prioritize the best evaluated resources
[r][c][
q
].avg
Q
Q
Quality_Metric
].min
Q[ ][c][ q ].maxQ (r,c) =
[ ][c][q
[ ][c][ q
].min
A Framework for Recommending Resource Allocation Based on Process Mining
26
Resource allocation metrics
Prioritize the cheapest resources
Cost_Metric
(r,c) =
Q[ ][c][co ].max
Q].avg
QQ[ ][c][co ].max
[r][c][co
[ ][c][co
].min
A Framework for Recommending Resource Allocation Based on Process Mining
27
Resource allocation metrics
Prioritize the most idle resources
[r][ ][ w ].top
Q
Q
Workload_Metric
].total
QQ[r][ ][ w ].top
(r,c) =
[r][ ][ w
[r][ ][ w
].bottom
A Framework for Recommending Resource Allocation Based on Process Mining
28
Resource allocation metrics
Prioritize the most experienced resources
logarithm
(
Q
Frequency_Metric
].total)
+1
logarithm
Q(
(r,c) =
].total) +1
A Framework for Recommending Resource Allocation Based on Process Mining
[r][c][ f
[ ][c][ f
29
Resource allocation metrics
Identify how below the resource is from the desired expertise
level
underQualification_ 1 1√∑ i =1
- n (under(i))2
Metric =
under(i
)=
Ec[i]-E r[i]if E c[i] >=E
Ec[i] - i r[i]
otherwi
0
se
A Framework for Recommending Resource Allocation Based on Process Mining
n
30
Resource allocation metrics
Identify how above the resource is from the desired expertise
level
overQualification_M 1 1√∑ i =1
- n (over(i))2
etric =
over(i)
=
Er[i]-E ifEr[i] >=E
c[i]
i E c[i]
0 - otherwi
c[i]
se
A Framework for Recommending Resource Allocation Based on Process Mining
n
31
Resource allocation request
1
2
Humanresources
information
Characteri
stics
Informati
on
Weights
ResourceAllocationRequest
Event log
3
Resource
ProcessCube
A Framework for Recommending Resource Allocation Based on Process Mining
32
Resource allocation request
Characterize
the request
Weights
Describe the importance of each dimension
Small size company:
Large size company:
F:010
P:050
F:025
P:015
Q:010
C:100
Q:100
C:030
U:015
O:000
U:075
O:065
A Framework for Recommending Resource Allocation Based on Process Mining
33
Proposed framework
1
2
Humanresources
information
Characteri
stics
Informati
on
Weights
ResourceAllocationRequest
Event log
3
Resource
ProcessCube
4
5
BPA
Best PositionAlgorithm
RecommenderSystem
A Framework for Recommending Resource Allocation Based on Process Mining
34
Implementation with BPA [10]
Position
Frequency
Performance
Quality
Data Item Local Score Data Item Local Score Data Item Local Score
1
R6
0.601
R3
0.851
R9
0.913
2
R4
0.593
R2
0.833
R7
0.864
3
R5
0.554
R1
0.832
R8
0.808
4
R1
0.525
R5
0.775
R2
0.723
…
…
…
…
…
…
…
…
…
… ……
…
…
…
…
… …
… …
… …
Overall top
Data item
Score
R1
0.795
R2
0.788
R14
0.784
[10] Akbarinia, R., Pacitti, E., Valduriez, P.: Best position algorithms for ecient top-k query processing, 2011
A Framework for Recommending Resource Allocation Based on Process Mining
35
Outline
Introduction
Approach
◦ General approach
◦ Resource allocation request
◦ Resource process cube and allocation metrics
◦ Implementation
Experimental evaluation
Future work
Conclusions
A Framework for Recommending Resource Allocation Based on Process Mining
36
Experimental evaluation
•
•
•
A Framework for Recommending Resource Allocation Based on Process Mining
37
Experiments results
Experiment 1
Top 3-queries
Single metric
A Framework for Recommending Resource Allocation Based on Process Mining
38
Experiments results
Experiment 2
Top 3-queries
3 help desk companies



A Framework for Recommending Resource Allocation Based on Process Mining
39
Experiments results
Experiment 3
Top 3-queries
3 help desk companies
Log size
Amount of resources
A Framework for Recommending Resource Allocation Based on Process Mining
40
Experiments results
( a ) Number of cases (thousands)
( b ) Number of resources
Performance analysis
A Framework for Recommending Resource Allocation Based on Process Mining
41
Outline
Introduction
Approach
◦ General approach
◦ Resource allocation request
◦ Resource process cube and allocation metrics
◦ Implementation
Experimental evaluation
Future work
Conclusions
A Framework for Recommending Resource Allocation Based on Process Mining
42
Future work
Explore potential application domains
Case studies using real data
Incorporate new dimensions
Combine our approach with existing works
A Framework for Recommending Resource Allocation Based on Process Mining
43
Outline
Introduction
Approach
◦ General approach
◦ Resource allocation request
◦ Resource process cube and allocation metrics
◦ Implementation
Experimental evaluation
Future work
Conclusions
A Framework for Recommending Resource Allocation Based on Process Mining
44
Conclusions
Consider different perspectives
Multi – factor criteria
New allocation technique
Sub-process level
Fine-grained, generic, & extensible
Experimental evaluation
BPA allows obtain a final ranking
Synthetic data
A Framework for Recommending Resource Allocation Based on Process Mining
45
THANK YOU FOR YOUR ATTENTION
A Framework for Recommending
Resource Allocation Based
on Process Mining
MICHAEL ARIAS
ERIC ROJAS
JORGE MUNOZ-GAMA
M A R C O S S E P Ú LV E D A
ANNEX 1
NEW DIMENSIONS
New dimensions
Incident management dimension
ITAT by priority
(Performance)
First Time Fix
(Quality)
# of incidents by priority
(Compliance)
Resolved within SLA by priority
(Value)
Problem management dimension
% Reduction incidents
(Value)
# of incidents resolved by…
(Quality)
# of proactive problems
(Compliance)
Average time to provide workaround by priority
(Efficiency)
A Framework for Recommending Resource Allocation Based on Process Mining
49
ANNEX 2
BEST POSITION ALGORITHM
Best Position Algorithm (BPA)
A Framework for Recommending Resource Allocation Based on Process Mining
51
BPA (iteration 1)
- SortedAccesstoPos1:
{d1,d2,d3}
- RandomAccess toobtain localscore yposition.
P11={1,4,9}and Bp1 =1
P12={1,6,8}and Bp2 =1
P13={1,5,8}yandBp3 =1
Bestpositionscore:
λ=f(S1 (1)+S2 (1)+S3 (1))
=30+28+30=88
A Framework for Recommending Resource Allocation Based on Process Mining
52
BPA (iteration 1)
L1 P1 = {1, 4, 9} and BP1 = 1
L2 P2 = {1, 6, 8} and BP2 = 1
L3 P3 = {1, 5, 8} and BP3 = 1
λ = f (S1(1), S2(1), S3(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
A Framework for Recommending Resource Allocation Based on Process Mining
53
BPA (iteration 2)
- SortedAccesstoPos2:
{d4,d5,d6}
- Random Access to obtain local score and
position.
P21={2,7,8}andBp21 =2
P22={2,4,9}andBp22 =2
P23={2,4,6}andBp23 =2
Bestpositionscore:
λ=f(S1 (2)+S2 (2)+S3 (2))
=28+27+29=84
A Framework for Recommending Resource Allocation Based on Process Mining
54
BPA (iteration 3)
- SortedAccesstoPos3:
{d9,d7,d8}
- RandomAccess toobtain localscore andposition.
P31={3,5,6}andBp31 =9
P32={3,5,7}andBp32 =9
P33={3,9,10}andBp33 =6
Bestpositionscore:
λ=f(S1 (9)+S2 (9)+S3 (6))
=11+13+19=43
A Framework for Recommending Resource Allocation Based on Process Mining
55
BPA (iteration 3)
P31 ={3, 5,6}Bp1=9,P32 ={3, 5,7}Bp2=9,P33 ={3, 9,10}Bp3=6
PL1 =PIt1L1UPIt2L1 UPIt3L1={1,2,3,4,5,6,7,8,9}
PL2 =PIt1L2UPIt2L2 UPIt3L2={1,2,3,4,5,6,7,8,9}
PL3 =PIt1L3UPIt2L3 UPIt3L3={1,2,3,4,5,6,8,9,10}
λ=f(S1(9),S2(9),S3(6))=11+13+19=43
LastGlobalScore:
{d3, d4,d6}={70,66, 70}
CurrentGlobalScore: {d9,d7,d8}={62,61, 71}
UpdateY. Y = {d3,d6,d8},= {70,70,71}
ThevaluesinYaregreaterthanλ= 43,thestopconditionfinish inposition3.
A Framework for Recommending Resource Allocation Based on Process Mining
56
BPA (iteration 3)
A Framework for Recommending Resource Allocation Based on Process Mining
57