process mining-driven optimization of a consumer loan approvals process
DESCRIPTION
Process Mining-Driven Optimization of a Consumer Loan Approvals Process. The BPIC 201 2 Challenge. Outline. 1 、 Introduction 2 、 Materials and Methods 3 、 Understanding the Process in Detail 4 、 Assessing Process Performance 5 、 Discussion 6 、 Conclusions 7 、 Homework. Introduction. - PowerPoint PPT PresentationTRANSCRIPT
1
Process Mining-Driven Optimization of a Consumer
Loan Approvals ProcessThe BPIC 2012 Challenge
2
Outline
1、 Introduction2、Materials and Methods3、 Understanding the Process in Detail4、 Assessing Process Performance5、 Discussion6、 Conclusions7、 Homework
3
Introduction
• In 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
4
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 (Delined).
5
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 trough 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.
6
Developing Thorough Understanding of the data
7
Tools used for analysis
• Disco– Preprocessing and exportation of data into formats suitable for
Microsoft Excel analysis.
• Microsoft Excel– Used Excel alongside Disco, which helped us visualize, rationalize
and refine observations in real time.
• CART Implementation from Salford Systems– Conducting preliminary segmentation analysis of the loan
applications to assess opportunities for prioritizing work effort.
8
Disco registration
• 使用下列網址下載 Disco:http://fluxicon.com/disco/
• Disco首頁進行 Mail註冊:
學校信箱
9
Setup Disco-1
點 Next
點Install
10
Setup Disco-2
點Finish
點 I accept
11
Setup Disco-3
學校信箱
點Registe
r
12
Setup Disco-4
學校信箱註冊碼
點Complet
e
13
Setup Disco-5
點 OK
14
起始畫面
開啟 log檔
15
載入 Log檔
選擇檔案
點開啟舊檔
16
主畫面 -1
17
主畫面 -2
開始檔案
過濾器 Case動畫 匯出 Log檔
18
Simplifying the Event Log
19
Simplifying the Event Log
O_CANCELLED 會與 A_CANCELLED及A_DECLINED同時發生。O_DECLINED
20
Filter
點此按鈕
選Attribu
te
選剔除掉的屬性按 Apply
21
過濾後畫面
匯出 CSV
22
匯出 CSV-1
選 Event log
選 CSV
選Export
23
匯出 CSV-2
點存檔
24
Simplifying the Event Log
25
Result
Activities 拉到
0%
26
Determining Standard Case Flow
27
Understanding Eventual Outcomes for Each Application
28
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
29
Case-Level Analysis
30
Case-Level Analysis
申請人往往會選擇一個 round number。 EX:5000、 10,000、 15,000
分成965、 966個 case來看
31
Case-Level Analysis
核准
取消
正在處裡
32
Performance of the Top 5 Resources based on Time Spent
最有經驗的人所花的時間 >平均各領域中花最多時間的五人
33
Leveraging Behavioral Data for Work Effort Prioritization
Salford Systems (http://www.salfordsystems.com)
Node 14:818 Case
Node 1:200 Case
34
Discussion
• Managing Event Complexity in Data– The event log would also benefit from
consolidation of events that happen concurrently, such as those that occur when successful applications are approved (A_APPROVED, A_REGISTERED and A_ACTIVATED).
35
Conclusions
• More extensive work in this area would be greatly aided by the inclusion of additional data points, such as customer information, policies that govern the process, operating costs for the process and eventual customer value.
• The bank would find significant additional benefits from exploring such additional areas, for example , social network analysis.
36
Homework
① 使用本篇銀行 log檔,依照您的觀點,提出”簡化流程”的方法。② 將簡化後的流程結果,匯出 JPG及轉存成
CSV檔。③ 繳交信箱: [email protected]④ 繳交日期: 2013/11/15(五 )
37
Thanks For Your Listening!!