Transcript process mining

流程探勘與它在評估流程
成本及分析流程上之應用
徐立群
成功大學會計學系
長榮大學資訊暨工程學院
Outline
 Introduction to Process Mining
 Data explosion challenge ; exploit the value of
ever increasing business data
 Process mining
 The three main purposes of process mining:
discovery, conformance, and enhancement
 Estimating process costs w/ process mining
 Analyzing processes w/ process mining
 An example taken from BPIC 2012 challenge
Physical universe vs. digital universe
 “The state of a bank” is mainly determined
by the data stored in the bank’s
information system. Money has become a
predominantly digital entity
.
 Nowadays, when the booking of an HSR
seat is successful, the customer can
receive an e-ticket that can be carried by a
smart phone.
Page 4
Physical universe vs. digital universe
(cont)
When the ERP system of a
manufacturer indicates that a
particular product is out of stock, it is
impossible to sell or ship the product
even when it is available in physical
form.
Technologies such as RFID, GPS, and
sensor networks will stimulate a
further alignment of the digital
universe and the physical universe.
Page 5
Exploiting the value of ever
increasing business data
One of the main challenges of today’s
organizations is to extract information
and value from data stored in their
information systems.
The importance of information systems
is not only reflected by the spectacular
growth of data, but also by the role
that these systems play in today’s
business processes as the digital
universe and the physical universe are
becoming more and more aligned.
Page 6
Exploiting the value of ever
increasing business data(cont)
 The growth of a digital universe that is well-aligned
with processes in organizations makes it possible
to record and analyze events.
 The challenge is to exploit event data in a
meaningful way, for example,






to provide insights,
identify bottlenecks,
anticipate problems,
record policy violations,
recommend countermeasures,
and streamline processes.
That is what process mining is
all about.
Page 7
Process mining vs BPM vs WFM
Process mining, i.e., extracting valuable,
process-related information from event
logs, complements existing approaches
to Business Process Management (BPM).
BPM is the discipline that combines
knowledge from information technology
and knowledge from management
sciences and applies this to operational
business processes.
BPM can be seen as an extension of
Workflow Management (WFM).
Page 8
Process mining vs BPM vs WFM (cont)
WFM primarily focuses on the
automation of business processes,
whereas BPM has a broader scope:
from process automation and process
analysis to process management and
the organization of work.
BPM aims to improve operational
business processes; for example, by
modeling a business process and
analyzing it using simulation,
management may get ideas on how to
reduce costs while improving service
levels.
Page 9
Example process model
A Petri net modeling the handling of compensation request (taken from
van der Aalst Process Mining text)
Page 10
Process Models --- Informal and
Formal
 Process models play important role in larger
organizations.
 Typically, two types of models are used: (a) informal
models and (b) formal models (also referred to as
“executable models”).
 Informal models are used for discussion and
documentation whereas formal models are used for
analysis or enactment (i.e., the actual execution of
process).
Process Models --- Informal and
Formal (cont)
 The value of models is limited if too little attention is
paid to the alignment of model and reality.
 For example, it makes no sense to conduct simulation
experiments while using a model that assumes an
idealized model---incorrect redesign decisions are
made.
 It is also precarious to start a new implementation
project guided by process models that hide reality.
Limitations
 Executable models may be used to force people
to work in a particular manner.
 However, most models are not well-aligned with
reality.
 Most hand-made models are disconnected from
reality and provide only an idealized view on the
processes at hand: “paper tigers”.
 Given (a) the interest in process models, (b) the
abundance of event data, and (c) the limited
quality of hand-made models, it seems
worthwhile to relate event data to process
models: process mining!
Process Mining
The idea of process mining is to
discover, monitor and improve real
processes (i.e., not assumed
processes) by extracting knowledge
from event logs readily available in
today’s systems.
Page 14
The three main uses of process mining:
discovery, conformance, and enhancement
Page 15
Estimating process costs w/
process mining
16
Event logs
(taken from van der Aalst Process Mining text)
航空公司賠償流程
以代號表示
Petri Net
Annotated transition system
Case 1
Case 5
Case 3
Case 6
Case 4
Case 2
Case id
Trace
1
abdeh
2
adceg
3
acdefbdeg
4
adbeh
5
acdefdcefcdeh
6
acdeg
Decide the probabilities
1/6
1
1
Case id
Trace
1
abdeh
2
adceg
3
acdefbdeg
4
adbeh
5
acdefdcefcdeh
6
acdeg
1
1/2
1
1
1
1
1
1
1
2/3
1/2
1
3/6
1
1
1
1
1
Activity
1/3
1/2
1
1
2/6
1/2
1
1
a
register request
b
examine throughly
c
examine casually
d
check ticket
e
decide
f
reinitiate request
g
pay compensation
h
reject request
Observations
1/6的Case會先
詳細的檢查(b) ,
但最後被拒絕
重新申請的Case
有1/2會被接受，
有1/ 2會被拒絕
3/6的Case會大
略的檢查(c) ->
1/3接受，2/3再
重新申請
✔
✔
2/6會先驗票(d)，
最後有1/2會被接受，
有1/2會被拒絕
流程拖的較長的
case都牽涉到
「大略檢查」
(activity c)
✔
整體來看，六個Case
的求償成功率為1/2。
（h = reject request）
Process cost
利用Event log 所留下的Cost,計算流程成本
Activity
Cost
a
register request
50
examine
throughly
400
c
examine
casually
400
d
check ticket
100
e
decide
200
Case id
Trace
Cost
1
2
3
4
abdeh
adceg
acdefbdeg
adbeh
acdefdcefcde
h
acdeg
SUM
AVG
950
950
1850
950
b
f
reinitiate
request
200
2750
g
pay
compensation
200
h
reject request
200
5
6
950
8400
1400
50+400+100+200+200
= 950
流程成本
$950
$2750
2300×⅔+950×⅓≅1850
$1850
$950
950×⅙+1850×½+950
×⅓≅1400
950×½+950×½≅950
$950
$950
在發生了acdef流程時，預
期此處流程的成本為$2300
2750×½+1850×½=2300
流程成本-決策樹
用決策樹的方式，不止可以得到平均成本＄1400,也可以得到
各個決策點的總流程成本。管理者可以思考如何改進流程以降
低總成本。
$950
$950
$2750
$2300
$1400
$1850
$2750
$1850
$950
$950
$1850
在發生了acdef流程後，預期整
$950
個流程的成本為$2300，高於
$950
$950 原來預期的1400
$950
$950
Analyzing processes w/ process
mining
 BPIC 2012 on the loan and overdraft approvals process of a
real-world financial institution in the Netherlands.
 Attempted to investigate following areas in detail：






Develop thorough understanding of the data
Develop a detailed understanding of the underlying process
Understand critical activities and decision points
Understand and map life cycle of a loan application from start to
eventual disposition as approved, declined or cancelled
Identify any resource level differences in performance one can discern
based on available data
Identify opportunities for “process interventions”：places in the
process based on likelihood of success
27
Materials and Methods
The data captures process events for
13,087 loan / overdraft applications over
a roughly six month period from October
2011 to March 2012.
The event log is comprised of a total of
262,200 events within these 13,087
cases.
Starting with a customer submitting an
application and ending with eventual
conclusion of that application into an
Approval, Cancellation or Rejection 28
(Delined).
Log Description
 An application is submitted through a webpage.
Then, some automatic checks are performed,
after which the application is complemented
with additional information.
 This information is obtained through contacting
the customer by phone. If an applicant is
eligible, an offer is sent to the client by mail.
 After this offer is received back, it is assessed.
When it is incomplete, missing information is
added by again contacting the customer.
 Then a final assessment is done, after which the
application is approved and activated.
29
Developing Thorough Understanding of the
data
30
Determining Standard Case Flow
31
Determining Standard Case Flow
Original
Explanation
afhandelen
leads
Following up on incomplete initial submissions
completeren
aanvraag
Completing pre-accepted applications
nabellen
offertes
Follow up after transmitting offers to qualified
applicants
nabellen
incomplete
Seeking additional information during
assessment phase
valideren
aanvraag
Assessing the application
32
Understanding Eventual Outcomes for Each
Application
33
Understanding Eventual Outcomes for Each
Application
1/4立即被拒絕
開始到後續大約剩下
1/4被拒絕
34
Case-Level Analysis
35
Case-Level Analysis
分965或966個
case為一個區
段來看
36
Case-Level Analysis
拒絕
取消
核准
37
Performance of the Top 5 Resources based on
Time Spent
最有經驗的人所花的
時間>全體平均
38
Conclusions
Process mining is a new technique that
allows one to zoom into the details of
actual business processes --- X ray for
business processes.
More applications are needed to show
its usefulness.
39
Thank you for your
listening!
Department of Accountancy
National Cheng Kung University
Chang Jung Christian University
College of Information & Engineering
40