Proactive process adaptation facilitates preventing or mit- igating upcoming problems during process execution, such as process delays. Key for proactive process adaptation is that adaptation decisions are based on accurate predictions of problems. Previous research focused on improving aggregate accuracy, such as precision or recall. However, aggregate accuracy provides little information about the error of an indi- vidual prediction. In contrast, so called reliability estimates provide such additional information. Previous work has shown that considering reli- ability estimates can improve decision making during proactive process adaptation and can lead to cost savings. So far, only constant cost func- tions have been considered. In practice, however, costs may differ depend- ing on the magnitude of the problem; e.g., a longer process delay may result in higher penalties. To capture different cost functions, we exploit numeric predictions computed from ensembles of regression models. We combine reliability estimates and predicted costs to quantify the risk of a problem, i.e., its probability and its severity. Proactive adaptations are triggered if risks are above a pre-defined threshold. A comparative eval- uation indicates that cost savings of up to 31%, with 14.8% savings on average, may be achieved by the risk-based approach.
%in kaalfontein+277-882-255-28 abortion pills for sale in kaalfontein
Risk-based Proactive Process Adaptation
1. Risk-based Proactive Process
Adaptation
Andreas Metzger, Philipp Bohn
Full Paper Presentation at the 1th International Conference on Service-Oriented
Computing - 15th , ICSOC 2017, Malaga, Spain, November 13-16, 2017, Lecture Notes
in Computer Science, E. M. Maximilien, A. Vallecillo, J. Wang, and M. Oriol, Eds., vol.
10601, Springer, 2017, pp. 351–366.
https://doi.org/10.1007/978-3-319-69035-3_25
(Open Access)
3. Motivation
Problem statement from a practical viewpoint
Domain: Freight Transport and Logistics
• Imagine a transfer hub,
e.g., an airport
3ICSOC 2017, Málaga
Point of
Prediction
RCS: Freight received from the shipper
DEP: Freight departed from airport
RCF: Freight received at arrival airport
DLV: Deliver freight from arrival airport
Make a decision to
mitigate
Challenge:
Deliver in time &
within reasonable
amount of costs
Problems occur:
e.g, delay in a previous
process instance
4. Motivation
Predictive Process Monitoring & Proactive Adaptation
4ICSOC 2017, Málaga
monitor
predict
real-time
decision
proactive
adaptation
time
t t +
planned /
acceptable situations
= Violation
= Non-
Violation
e.g., delay in
freight delivery
time
e.g., schedule
faster means of
transport
5. • Error of the individual prediction
• Prediction may be wrong
Prediction reliability to support decision [Metzger and Föcker 2017]
• Cost savings through adaptation
• E.g., costs imposed by contractual penalties
such as stipulated in SLAs
Factoring in cost model as additional decision support
ICSOC 2017, Málaga 5
Motivation
Factors impacting the success of the adaptation decision
Contribution
6. Motivation
Risk-based Proactive Process Adaptation Decision
6ICSOC 2017, Málaga
monitor
predict
time
t t +
planned /
acceptable situations
= Violation
= Non-
Violation
R ≤ threshold no adaptation
R > threshold adaptation
+ Risk R
real-time
decision
proactive
adaptation
Contribution
7. ICSOC 2017, Málaga 7
Motivation
Solution Idea
• Risk = probability of occurrence × severity
as in ISO 31000:2009 [Purdy 2010]
• Risk = Reliability estimate × Penalty
11. 11ICSOC 2017, Málaga
δ
Constant
Step-wise (s steps)
δ
Linear with cap
clin
c
0
cconst
c
δ1/s 2/s
cstep
1
c
2/s·cstep
1/s·cstep
(s-1)/s
…
1 0 1
0
Evaluation
Shapes for cost model
Contribution
12. Evaluation
Independent Variables
• Penalty cost functions serving two purposes:
1. Compute predicted penalty Severity of the risk
2. Compute actual penalty according to cost model
• Adaptation effectiveness
• If adaptation helps to achieve objective, it is considered as effective
• Added as probability indicating the effectiveness of adaptation
• Risk threshold
• Decides if an adaptation is triggered
• Used to reflect different attitudes towards process risks
Except of risk threshold all variable are given in a concrete
problem situation.
12ICSOC 2017, Málaga
13. Evaluation
Process Model and Data Set
Domain: Freight Transport and Logistics
• Airfreight process
• 5 months of operational data
• 3 942 process instances
• 56 082 service invocations
13ICSOC 2017, Málaga
Point of
Prediction
14. Evaluation
Effect on Costs (compared with reliability-based)
Penalty R = 0.1 R = 0.3 R = 0.5 R = 0.7 R = 0.9
constant -19.0 -20.0 -17.0 -3.0 3.1
step-wise -14.0 12.0 20.0 20.0 8.6
linear 0.6 21.0 27.0 26.0 11.0
ICSOC 2017, Málaga 14
Averaged over probability of effective process adaptation (α) ={0.1, 0.2, 0.3, … ,1}
Constant penalty Step-wise penalty Linear penalty
Risk threshold R
Costsavings
16. Conclusions
Observation
• Cost savings for risk-based approach on SLA cost model
• But also on adaptation cost model. See our paper
• Savings as high as 31.0%
• Average savings of 14.8%
• Average savings of 23.4% for non-constant cost models
Future work
• Considering aggregate SLAs
• Replication with data from
port terminal operations
16ICSOC 2017, Málaga
17. Thanks
ICSOC 2017, Málaga 17
…the EFRE co-financed operational
program NRW.Ziel2
http://www.lofip.de
…the EU’s Horizon 2020 research and
innovation programme under Objective
ICT-15 ‘Big Data PPP: Large Scale Pilot
Actions ‘
http://www.transformingtransport.eu
Research leading to these results has received
funding from…
18. References
[Metzger and Föcker 2017] Metzger, A., Föcker, F.: Predictive business process
monitoring considering reliability estimates. In: Dubois, E.,
Pohl, K. (eds.) 29th Intl Conference on Advanced
Information Systems Engineering (CAiSE 2017), Essen,
Germany. LNCS, vol. 10253. Springer (2017)
[Purdy 2010] Purdy, G.: ISO 31000:2009 – setting a new standard for risk
management. Risk analysis 30(6), 881–886 (2010)
ICSOC 2017, Málaga 18