This document discusses challenges with embedding decisions in enterprise applications and outlines an approach called Decision Service Management. Some key points:
1) The traditional waterfall methodology does not work well for decisions since they require ongoing optimization after go-live as factors change. Decisions are also high-change components that need to be adaptable.
2) Embedding decisions directly in processes can lead to complex, rigid processes that are difficult to change. Decisions are better managed separately from processes.
3) Decision Service Management separates decisions from processes and treats them as first-class citizens. This allows for continuous improvement of decisions and more agility. BRFplus is highlighted as a tool that supports this approach.
Report
Share
Report
Share
1 of 47
More Related Content
Building Flexible, Easy to Change, and Rock-Solid Applications with BRFplus Decision Services
1. Building flexible, easy to change and rock-
[ solid applications with BRFplus decision
services
Carsten Ziegler, James Taylor
2. [ Learning Points
Learn how the empowerment of business experts is built
into SAP applications
See how to achieve continuous decision and process
improvement
Get step-by-step instructions on how to apply this
approach in your systems
Real Experience. Real Advantage. 2
3. [ Presenters
Carsten Ziegler – SAP
SAP-wide Business Rules Architect
Chief Product Owner BRFplus
James Taylor – Decision Management Solutions
CEO, Decision Management Solutions
Leading expert in Decision Management and
the application of Business Rules
Real Experience. Real Advantage. 3
4. [ Agenda
Presenters
Challenges in Enterprise Systems
The Decision Service Management Approach
Decision Service Management with BRFplus
ROI
Best Practices and Learnings
Real Experience. Real Advantage. 4
5. [
“ Once you pour electronic concrete,
it’s hard to get out. ”
Shell
“ There is no such thing as a
temporary solution. Don’t put
in things you don’t want to ”
keep.
Real Experience. Real Advantage.
6. [ Enterprise Applications
Report
but don’t learn
Wait
rather than act
Local
exceptions
Global
standards Built to last,
Escalate rather not to change
than empower
Real Experience. Real Advantage. 6
7. [ To improve they must make more decisions
Manual Not Made
Before Managed
After Not Made
Manual
Managed
New
Reproduced from Smart (Enough) Systems, Prentice Hall 2007 Larger boxes represent more decisions, by volume
Real Experience. Real Advantage. 7
8. [ Decisions
Making choices
or selections and
taking action
Real Experience. Real Advantage. 8
9. [ Candidate decisions
Determine if a customer is eligible for a benefit
Validate the completeness of an invoice
Calculate the discount for an order
Assess the risk of a transaction
Select the terms for a loan
Choose which claims to Fast Track
These are decision words
The system must answer a question each time
Real Experience. Real Advantage. 9
10. [ Problem 1: Waterfall does not work for decisions
Standard methodology for implementation and change projects
Assumes a phased/waterfall approach:
Blueprint > Realization > Testing > Productive use
Problems
Assumes you know everything when you start, though you don’t
No structure for ongoing optimization after go-live
What you learn in realization cannot be fed back into your blueprint
(particularly bad for decision-making!)
Real Experience. Real Advantage. 10
11. [ Decisions are high change components
Regulations change Change to keep eligibility decisions compliant
Policies change Change validation to track new data
requirements
Competitors change Change the discount to remain competitive
Markets change Change the assessment to manage risk
Consumer behavior changes Change to keep selecting the right
deal terms
Fraudsters adapt Change the routing to focus on new fraud
Real Experience. Real Advantage. 11
12. [ Automation vs. Agility or Business vs. IT
Automation: Full Semi No
Business
Expert
Business Business
Expert Analyst
Hard-coded decisions Decisions in documents Decisions as tacit knowledge
Agility: Low Medium High
Questions
What are my costs for changes?
How to gain transparency for business users?
How to enforce policies in business operations?
How do operations scale?
How to ensure process quality?
How to achieve legal compliance?
Real Experience. Real Advantage. 12
13. [ Problem 2: Decisions, and rules, are not processes
Age<21
Accept low-risk
applicant
21<=Age<50
Process
medium-risk
applicant
Age>50
Decline high-
risk applicant
Real Experience. Real Advantage. 13
14. [ Problem 2: Decisions, and rules, are not processes
Age<21
Accept low-risk
applicant
Previous Heart Attack,
Existing Cancer Decline high-
risk applicant
21<=Age<50
Process
medium-risk
Good Medical Record applicant
Age>50
Real Experience. Real Advantage. 14
15. [ Problem 2: Decisions, and rules, are not processes
Age<21
Accept low-risk
applicant
Previous Heart Attack,
Existing Cancer Decline high-
risk applicant
21<=Age<50
Process
medium-risk
Good Medical Record applicant
Age>50 1 or fewer Claims Multiple Claims
New Customer
Long-standing Customer
Real Experience. Real Advantage. 15
16. [ Problem 2: Decisions, and rules, are not processes
Age<21 No Class I or II sports
Accept low-risk
applicant
Previous Heart Attack, Class II sport
Existing Cancer Decline high-
risk applicant
21<=Age<50
Class I sport
Process
medium-risk
Good Medical Record applicant
Age>50 1 or fewer Claims Multiple Claims
New Customer
Long-standing Customer
Real Experience. Real Advantage. 16
17. [ Embedding decisions cause process complexity
Age<21 No Class I or II sports
Accept low-risk
applicant
Class II sport
Previous Heart Attack, Decline high-
Existing Cancer risk applicant
21<=Age<50
Class I sport
Process
medium-risk
Good Medical Record applicant
Age>50 1 or fewer Claims Multiple Claims
New Customer
Long-standing Customer
Real Experience. Real Advantage. 17
18. [ Separate decision, simpler process
Low risk
Process low-
risk applicant
Medium risk
Process
Determine
medium-risk
applicant type applicant
High risk
Process high-
risk applicant
Class I Class II New #
Age sports sports Heart Attack? Cancer? Customer? Claims Action
<21 N N Low Risk
<21 Y High Risk
<21 Y N Medium Risk
21 - 50 Y High Risk
21 - 50 Y High Risk
21 - 50 N N Medium Risk
>50 Y High Risk
>50 N <2 Medium Risk
>50 N >1 High Risk
Real Experience. Real Advantage. 18
19. [ Process rules are not decision rules
Flow control,
routing
IT ownership Business
Governed as part Process
of the process Rules
Lifecycle
dependent on
Business
process lifecycle
Process Decision Point
Tasks
Business decision
making
Business-side
ownership
Decision Rules Governed by the
businessGladys Lam, BRSolutions
After
Real Experience. Real Advantage. Independent lifecycle 19
21. [ Problem 3: Change and Transport Management
(CTS)
Development Quality Assurance Productive
System (DEV) System (QAS) System (PROD)
CTS is a set of tools and a methodology
implement changes in DEV, transport changes to QAS, and upon success transport to
PROD
CTS supports transport of customizing and repository objects
code, database/dictionary definitions
CTS locks customizing and repository from changes in PROD
Problem - one size does not fit all
Downtime required
CTS and working in the development system requires technical support
Master data needs to be replicated into DEV
Mandatory artificial classification of all artifacts into Customizing , repository
and application data (master data and transactional data) with strict change
management; no changes of customizing in PROD
Real Experience. Real Advantage. 21
22. [ Missed Opportunities
Missed Opportunities
Open Change Requests
Go-Live Initial Scope Go-Live Delta Scope
Traditional Approach
Waterfall process with release cycles: spec, design, code
Same change management for all objects (DB, UI, decision logic)
High costs also for small changes
Questions
How long will it take to get a requirement implemented when the cut-off
date passed by?
Will there be another project for my requirements in the future?
What will be the costs of the project and who will have to pay for it?
What will happen if the implemented decisions are outdated?
Real Experience. Real Advantage. 22
23. [ Agenda
Presenters
Challenges in Enterprise Systems
The Decision Service Management Approach
Decision Service Management with BRFplus
ROI
Best Practices and Learnings
Real Experience. Real Advantage. 23
24. [ Suitable Decisions
Banking: Relationship based Pricing, Education: Fee Calculations, Healthcare: Claims, Patient
Scorecards, Credit Decisioning Course Selections Monitoring, Fraud Detection
Logistics and Shipping: Public Sector: Tax Calculations, Insurance: New Products,
Parts Management, Duties Calculations, Customs Duties, Land Regulations, Claims Settlement, Agent Commissions
Pricing Calculations License Fee Calculations
Real Experience. Real Advantage. 24
25. [ These decisions involve many kinds of rules…
Decision
Real Experience. Real Advantage. 25
26. [ …none of which are manageable
public class Application {
private Customer customers[];
private Customer goldCustomers[];
...
public void checkOrder() {
for (int i = 0; i < numCustomers; i++) {
Customer aCustomer = customers[i];
if (aCustomer.checkIfGold()) {
numGoldCustomers++;
goldCustomers[numGoldCustomers] = aCustomer;
if (aCustomer.getCurrentOrder().getAmount() > 100000)
aCustomer.setSpecialDiscount (0.05);
}
}
}
Real Experience. Real Advantage. 26
27. [ Alternatives to business rules
Explicitly coded within application or database
Pros Cons
• Familiar to IT • IT department must be involved
in any change
• Simplifies development and • Little re-use of logic across
testing processes multiple decisions
Parameterized in tables or configuration files
Pros Cons
• Decreases cost of change • Use of parameters can decrease
readability
• No business/IT collaboration • Assumes kinds of changes can
be exhaustively specified
Real Experience. Real Advantage. 27
28. [ A BRMS like BRFplus is compelling because
Design Transparency
Execution Transparency
Agile Compliance
Collaboration
Platform for analytics
Formula
Real Experience. Real Advantage. 28
29. [ From Missed to Realized Opportunities
Realized Opportunities
Changes implemented
with decision services
Go-Live Initial Scope Go-Live Delta Scope
Decision Service Management Approach
Clear separation of decisions from processes (agile but robust systems)
Domain experts own and change implemented decisions
Permanent and incremental decision optimizations
Real Experience. Real Advantage. 29
30. [ Agenda
Presenters
Challenges in Enterprise Systems
The Decision Service Management Approach
Decision Service Management with BRFplus
ROI
Best Practices and Learnings
Real Experience. Real Advantage. 30
31. [ Decision Service Approach: Customer Landscape
Managed System
Connector
1
e.g. Business Suite Decision Service
Manager
Managed System
Connector
2
e.g. CRM
Decision Service Manager
Extremely low hardware requirements
Requires NetWeaver 7.0 Enhancement Package 3 or NetWeaver 7.3
Enhancement Package 1
Managed Systems
Existing servers with customer business applications
No Upgrade required, works with NetWeaver 6.40 or higher (>6 years
old)
Any number of managed systems can be connected with the Decision
Service Manager (RFC)
Real Experience. Real Advantage.
32. [ Decision Service Approach: Modeling
Managed System
Connector
1
Decision Service Manager
Managed System
Connector
2
Decision Service Modeling with BRFplus
Central creation, test and management of all rules and decisioning logic
Service implemented in rulesets using decision tables, trees, text rules,
formulas
Transparency - no code but graphical modeling
Access to all relevant data in the managed systems (e.g. master data
used in value help of decision table rules)
Real Experience. Real Advantage. 32
33. [ Decision Service Approach: Deployment
Managed System
Connector
1
Decision Service
Manager
Managed System
Connector
2
Decision Service Deployment
Deployment describes the transfer of the decision service into one or
many managed systems
Hot deployment - immediate use with no downtime for deployment
Activity periods - deploy today with as-of usage timepoint
Overview about all decision services in all systems, deployment
statistics
Real Experience. Real Advantage. 33
34. [ Decision Service Approach: Use
Managed
Connector
System 1
Decision Service
Manager
Managed System
Connector
2
Decision Service Use
Local execution of generated code, mass-enabled
Best performance
No dependency on Decision Service Manager
Execution traces (inputs, result, decision path for decisioning analytics
or legal compliance)
Real Experience. Real Advantage. 34
35. [ Decision Service Approach: Continuous
Improvement
Managed System
Connector
1
Decision Service
Manager
Managed System
Connector
2 Optimize
Decision Service Optimization
Super-fast change cycles in the hand of the domain experts:
Analyze, Optimize, Implement
Automated business decisions with full transparency for better decision
quality
Immediate ROI
Real Experience. Real Advantage. 35
36. [ Decision Service Approach BW
Connector
ERP
Connector
Decision Service
Non-
Manager Web Service
SAP
CRM
Connector
Custom
Connector
SRM App
Connector
Decision Service enabled customer system landscape
Deployment of one service into multiple systems
Reuse of decision services and single rules, decision tables, etc.
Non-SAP systems connected by web service
Real Experience. Real Advantage. 36
37. [ Agenda
Presenters
Challenges in Enterprise Systems
The Decision Service Management Approach
Decision Service Management with BRFplus
ROI
Best Practices and Learnings
Real Experience. Real Advantage. 37
38. [ The ROI of Decisions and Business Rules
1. Increased Business Value
2. Increased Development Efficiency
3. Increased Maintenance Efficiency
And in that order!
Real Experience. Real Advantage. 38
39. [ Business Value
Reduced losses from more accurate fraud decisions
Reduced fines from demonstrable compliance with
regulations
Quicker time to market with new products or services
More competitive pricing scheme
Productivity of business staff increases
Redeployment of assets when staff no longer need to do
repetitive decision-making
Reduced time to value for new systems functionality with
benefits of a system being realized sooner
Reduced costs or losses while waiting for a change to a
system such as rework caused by out of date system
behavior.
Real Experience. Real Advantage. 39
40. [ Development Efficiency
Increased specification productivity as business analysts
are involved directly
Improved developer productivity—adding 1 business
analyst can eliminate the need for multiple programmers
Reduced re-work as there is an improved quality of
business rules
Reuse of previously developed logic because it is easier
to see opportunities for reuse and to take advantage of
them when logic is in rules in a repository
Real Experience. Real Advantage. 40
41. [ Maintenance Efficiency
Reduced effort to find where to make a required change
Reduced effort to understand the existing code so the
right change can be made
Reduced effort to make the change
Reduced effort to test the change to confirm it works
Reduced effort to deploy the change
Real Experience. Real Advantage. 41
42. [ Agenda
Presenters
Challenges in Enterprise Systems
The Decision Service Management Approach
Decision Service Management with BRFplus
ROI
Best Practices and Learnings
Real Experience. Real Advantage. 42
43. [ Best Practice: Begin with the Decision in mind
Find the decisions that
matter to your business
and focus on them as well
as your processes.
Real Experience. Real Advantage. 43
44. [ Best Practice: Business Ownership
Decisions have owners.
So have decision services.
Decision services are the
responsibility of business-side
experts.
Real Experience. Real Advantage. 44
45. [ Best Practice: 3 Steps to Decision Management
Discover
Build
Improve
Real Experience. Real Advantage. 45
46. [ Key Learnings
Focusing on Decisions improves Enterprise Applications
Decisions are different and not the same as Processes or
Rules
Managing Decision Services creates agility and business
value
BRFplus delivers Decision Services optimized for
SAP/ABAP applications
Use BRFplus Decision Services whenever one of the
following is true:
Decisions are value drivers
Decisions are likely to change over time
Configuration with customizing tables will become too
complex
Real Experience. Real Advantage. 46
47. [
Thank you for participating.
Please remember to complete and return your
evaluation form following this session.
For ongoing education on this area of focus, visit the
Year-Round Community page at www.asug.com/yrc
]
[ SESSION CODE:
2414
Real Experience. Real Advantage. 47
Editor's Notes
Big systems implementations tend to be “electronic concrete”. In particular you can’t easily change the logic in those systems. You can add process management across and within them, but many business processes are actually pretty stable – it is the decisions within them that are constantly changing. For instance, your order to cash process does not change but your pricing decision does. Identifying the decisions in your applications and processes explicitly and then automating them using business rules allows you to develop a system that is both robust and flexible.IT systems are notorious for lasting longer than expected. They rapidly evolve from “temporary solutions” to permanent ones. Those who think business rules are overkill often tell me that “this code isn’t going to change” and “it will always be easy for someone to change this code” despite decades of evidence that this is not the case. Instead of assuming it won’t change or that making changes will be easy, companies would do better assuming that they will need to change the logic in their systems more often than they think and that the people making the changes won’t be those who wrote it originally. The ease with which business rules can be changed and re-deployed and the greater clarity of business rules syntax make this work. Assuming that your systems will be permanent, and making sure that critical flex points in the system are easy to change (because they are externalized as rules-based decision services) will ensure that your have permanence with flexibility.
Decisions involve choices, selecting from alternatives and are critical to how we run our organizationsBetter decisions means making more effective, more profitable choices from those available – choices that improve what will happen in the future
much of a typical mainframe system is static, works fine and needs no maintenance. Often only a small portion of the system is responsible for much of the maintenance work. My suggestion would be to find the COBOL that represents business decisions such as a pricing engine (what price is this product for this customer), eligibility logic (is this customer eligible for this offer or service), approval rules (can this claim be auto-approved) and replace those parts of the application with Decision Services.
Lack of centralization (difficult reuse) + customers avoid updates (because of applicability for complete server with application stack -> retests) -> stuck in older technology
So making decisions correctly will be hard unless we can pull all these rules together.Given this is how rules often look to start with, this is clearly going to be hard.But rules also change…Because your business policies doBecause your competitors doBecause the law doesBecause stock levels doBecause your services and products doBecause your customers doBecause your customers’ needs doSo we need something that will let us collect, manage and update the business rules that drive our decisions
Increased business value is the largest and most compelling business benefit that can be realized from the use of a business rules management system such as BRFplus. These benefits are driven by improved effectiveness and efficiency in decision-making and by improved responsiveness as BRFplus frees business users from the tyranny of the IT work queue for changes. One company developed a BRFplus based system to determine appropriate carriers for 25,000 shipments per hour, replacing the emailing and management of large numbers of Excel spreadsheets.
Find the decisions that matter to your business and understand themDon’t start with your data, start with the decisions you need to improveStart with the definition of decision services - purpose and service interface (facts required)-> needssupportfrom IT (wheretoputtheservicecall)
Discover and Model DecisionsDesign and Implement Decision ServicesMonitor and Improve Decisions