This slide deck walks you through the basis of understanding prescriptive analytics. Understand the different kinds of prescriptive analytics, how it works, its value, where to find use cases and more!

1. Prescriptive Analytics
A Primer

2. Agenda
§ What is Prescriptive Analytics?
§ How does it work?
§ Where is it applied?
§ What is the value proposition of Prescriptive Analytics?
§ How do you identify the right use cases?
§ What are the technical implementation requirements?
§ Where to go next?

3. What is Prescriptive Analytics?
Understanding where Advanced Analytics Stand Today
• What happened?
• What is driving it?

4. Gartner’s Definition
“Prescriptive analytics is the application of logic and
mathematics to data to specify a preferred course of action.
While all types of analytics ultimately support better decision
making, prescriptive analytics outputs a decision rather than a
report, statistic, probability or estimate of future outcomes.”
-Gartner, “Forecast Snapshot: Prescriptive Analytics, Worldwide, 2015
Gartner defines two types of prescriptive analytics
§ Optimization
§ Heuristics

5. Role of Optimization in Prescriptive Analytics
§ Optimization solvers use algorithms such as linear programming, mixed
integer programming, constraint programming, and heuristic algorithms to
minimize or maximize some objective while meeting global business
constraints.
§ Optimization can tackle more complex problems by making use of an explicit
analytical decision model to compute the outcomes of each alternative and
evaluating trade-offs among multiple objectives and constraints.
§ This analytical decision model explicitly computes outcomes for each
decision alternative, so one knows for every possible choice what the
expected outcomes are and, in the end, what the optimal decision is.
§ In other words, Prescriptive Analytics is enabled by Optimization

6. § Heuristics- (or rules-) based modeling
o Logic is defined by users
o Works in sequence
o Difficult to implement advanced
constraints (i.e. throughput, min. batch
production)
§ Leads to “an answer” that may or may
not be feasible
§ Takes effort to maintain the rules (not just
the data) as conditions change (e.g.
product/services mix, throughput)
§ Provides limited insight beyond the
answer
“An answer” or “best possible”
Heuristics or Optimization
§ Optimization modeling
o Creates a “map” that represents reality;
i.e., given limits for sales, logistics,
production, stocking…
o Answers “what combination of activity is
best?”
o Flexibly maximizes/minimizes objectives
(i.e. profit, service level, cost)
§ Leads to the “best possible” answer
o Typical improvement of 10-20% over
heuristics based modeling
o By definition also doable
§ Provides additional insight into marginal
opportunities around demand and
business constraints

7. Heuristic/Rules Approach
$100 $55 $80
$65 $85
$110 $60 $30 $95
Rule - Use Line 1 to 80%
overflow to line 2
Rule – Check inventory, if
available make in two
weeks, else next two days
Rule – If line 1 use
method A, if line 2,
method B …
Where to Make Order?
When to Make
Order?
How to Make
Order?
Profit of Order
Ignores other possibilities
Answer may be infeasible
Good quick answer

8. Optimization Approach
$100 $55 $80
$65 $85
$110 $60 $30 $95
What’s the answer that
maximizes profit and is
also doable?
How do we get there
(prescription)?
On Line 5
Process today
Using method 10
Best feasible answer

9. Prescriptive Analytics Vs Hypothesis Driven Approaches (e.g.,
heuristics): Overview
Steps for Decision Making Hypothesis Driven Approach Prescriptive Analytics
Process Flow High-level representation of the problem
usually white-boarded. A hypothesis for
the problem under consideration is
documented at this step that drives the
rest of the steps.
Detailed representation of the problem
captured in business expert system. No
pre-conceived notion of a hypothesis is
required.
Decision Flow Only a representative set of decisions are
considered driven by the hypothesis.
Significantly larger number of decision
steps are considered due to the visual
nature of the model and ease of
modeling.
Data Sample data set that is relevant to the
hypothesis.
Larger data set that populates the scope
of the model and decisions under
consideration.
Constraints Not Applicable. Upper and lower control limits on the
key constraints that are modeled and
relevant to the decisions.
Prescription Not Applicable Model solves for different scenarios
drives the analysis of which decisions
results in highest net income.

10. Agenda
§ What is Prescriptive Analytics?
§ How does it work?
§ Where is it applied?
§ What is the value proposition of Prescriptive Analytics?
§ How do you identify the right use cases?
§ What are the technical implementation requirements?
§ Where to go next?

11. Steps for Prescriptive Analytics
§ Process & Decision Flow
§ Represent the process flow model for the organization (resources, costs, capabilities)
§ Determine the range of possibilities for decisions made by the organization
§ Data
§ Input data from representative historical period to feed the model
§ Validate the process flow
§ Constraints
§ Define the upper constraint and lower constraint limits for all decision possibilities
§ Establish the objective function(s) to be considered
§ Prescription
§ Identify the best way to utilize the resources, costs, and capabilities of the organization that
maximizes the defined objective(s)
§ Evaluate what-if scenarios

12. Prescriptive Analytics Drives Decision Making

13. Agenda
§ What is Prescriptive Analytics?
§ How does it work?
§ Where is it applied?
§ What is the value proposition of Prescriptive Analytics?
§ How do you identify the right use cases?
§ What are the technical implementation requirements?
§ Where to go next?

14. Where is Prescriptive Analytics Applied?
Industry Solution
Financial Services Cash Management
Financial Services Mortgage Services Strategy & Portfolio Optimization
Aerospace & Defense Service Contract Profitability Modeling
Healthcare - Providers Population Management & ACO Transition
Healthcare - Providers Staff, service, and resource optimization
Health Plans Health Plan Benefit Design Optimization
Health Plans Provider Network Optimization
Utilities Multiple – Strategic, Tactical & Operational in Water Utility Industry
Consumer Packaged Goods Trade Promotion Optimization (TPO)
Consumer Packaged Goods Integrated Business Planning / S&OP
Oil & Gas Logistics Optimization
Transportation & Hospitality Revenue Management & Logistics Optimization
Retail Price & Promotions Optimization
High Tech Integrated Business Planning (S&OP)
Chemicals Integrated Business Planning (S&OP) / Capex
Government Army recruiting
Natural Resources Network Optimization / Integrated Business Planning (S&OP) / Capex
Metals Product Mix & Supply Planning
Mining Supply Chain Planning & Blend Optimization
Note: Certain types of
problems are naturally
optimization problems, while
others call for a simpler
solution, or one that builds on
existing predictions and rules.

15. Agenda
§ What is Prescriptive Analytics?
§ How does it work?
§ Where is it applied?
§ What is the value proposition of Prescriptive Analytics?
§ How do you identify the right use cases?
§ What are the technical implementation requirements?
§ Where to go next?

16. Typical Value Realized by Prescriptive Analytics
§ Understand true baseline performance potential
§ Optimize performance across operational, tactical and strategic applications
§ Operational use cases (i.e., in-patient department optimization, truck optimization)
typically deliver 15-30% improvement in cost, throughput and other key metrics
§ Enterprise-wide use cases (i.e. integrated business planning, revenue management) deliver
up to 8% of annual revenue in additional profitability, as well as revenue growth
opportunities
§ Establish dynamic understanding of average, marginal cost and profitability,
including products, customers and constraints
§ Improve financial and operational forecast accuracy
§ Capture and expand organizational knowledge about the business

17. Agenda
§ What is Prescriptive Analytics?
§ How does it work?
§ Where is it applied?
§ What is the value proposition of Prescriptive Analytics?
§ How do you identify the right use cases?
§ What are the technical implementation requirements?
§ Where to go next?

18. How do you Identify the Right Use Cases?
§ Policies that guide behavior, where decisions are made out of habit or where
people fail to make decisions
§ E.g., we always source products from this plant to serve this market; we prioritize our
customers based on volume/revenue
§ E.g., all our surgeries start at 6am because that’s how we’ve always done it; our
oncology clinic opens from 8am-5pm, even though we’ve never really analyzed why
§ On-going complex planning processes that are treated sequentially today or,
even worse, that are made in isolation. Examples include:
§ Sales & operations planning, where decisions about sales/marketing, manufacturing,
procurement, distribution and finance are made sequentially using different tools
and/or spreadsheet models (Note: S&OP processes exist in almost every industry)
§ Decisions driven by silo thinking — tactical and strategic decisions involving
resources, product/service mix, marketing, etc. — that are made solely within the
function

19. How do you Identify the Right Use Cases? (Continued)
§ Highly dynamic situations where input/product prices change constantly,
regulations evolve, etc. – these situations require dynamic optimization
analyses
§ Examples include commodities industries, the U.S. healthcare industry, chemicals, oil
& gas, some finance products, etc.
§ High difference in average vs. marginal decision making. When there are
multiple constraints, volume contracts and output price differentials, very
often there can be up to 100% difference in average vs. marginal profitability
for the same products
§ Underserved markets/industries – markets where the problems are relatively
complex, but where practitioners over-rely on BI and Excel tools to make
decisions. Examples include some government services, some US healthcare
providers, etc.

20. Agenda
§ What is Prescriptive Analytics?
§ How does it work?
§ Where is it applied?
§ Value proposition of Prescriptive Analytics
§ How do you identify the right use cases?
§ What are the technical implementation requirements?
§ Where to go next?

21. What Are the Business Requirements?
§ Compelling ROI – clear value add over and above prior alternatives
§ Great user experience – easy to learn, easy to use
§ Develop, maintain and extend models
§ Input, prepare, visualize and edit input data
§ Visualize and compare scenarios
§ Ability for key users to configure their own reports
§ Trustworthy business process – explicit, collaborative
§ Input validation
§ Process tracking & auditing
§ User collaboration
§ Agility – intelligent, fast
§ Proactively identify problems, including comparison of actuals vs. plan
§ What-if scenario analysis

22. Key Technical Requirements for Successful Implementation
§ Visual optimization modeling with expert knowledge embedded
§ Quickly create, maintain and extend prescriptive analytics models
§ Provide intelligent recommendations that tie to business value
§ Highlight additional opportunities and risks
§ User interface that provides a compelling user experience
§ Data aggregation/disaggregation logic with exception management
§ Best practice scenario management & visualization
§ Ad-hoc, user-configurable reporting
§ Multi-user/enterprise wide platform
§ Robust and scalable data management capabilities
§ Workflow management
§ Cloud based including key services (e.g. single sign-on, security, etc.)
§ Ability to support operational as well as strategic planning processes

23. Agenda
§ What is Prescriptive Analytics?
§ How does it work?
§ Where is it applied?
§ What is the value proposition of Prescriptive Analytics?
§ How do you identify the right use cases?
§ What are the technical implementation requirements?
§ Where to go next?

24. Where to Go Next?
§ Understand the market space
§ Gartner’s Market Guide to Optimization Solutions
§ Gartner’s Forecast Snapshot: Prescriptive Analytics Worldwide, 2015
§ Sample Case Studies for Finance
§ Sample Case Study in CPG
§ Brainstorm use cases and potential value add
§ Ensure participation by business SMEs
§ Use River Logic’s Use Case Definition Framework as a guide
§ Assess organizational readiness
§ Asses maturity of advanced analytics using Gartner’s maturity curves
§ Asses maturity of S&OP Process and SCP using Gartner’s maturity curves
§ Contact the experts
§ River Logic
§ River Logic Partners