Module 5: Prescriptive Analytics
What is Prescriptive Analytics?
1. Prescriptive Analytics is the final stage of the business analytics
process, following descriptive and predictive analytics.
2. While descriptive analytics explains what happened and predictive
analytics forecasts what is likely to happen, prescriptive analytics
recommends actions to achieve specific business objectives.
3. It uses advanced techniques like optimization, simulation, decision
analysis, and machine learning to prescribe the best course of
action among various options.
💡 Definition:
Prescriptive analytics provides data-driven recommendations for decision-
making by identifying optimal solutions or strategies under given constraints.
Importance of Prescriptive Analytics
1. Helps organizations make better and faster decisions.
2. Bridges the gap between data analysis and business strategy.
3. Offers a competitive advantage by recommending actions to maximize
profits or minimize costs.
4. Valuable in areas such as supply chain optimization, pricing
strategies, production planning, and resource allocation.
Techniques in Prescriptive Analytics
✅ A. Optimization Techniques
✅ B. Simulation Modelling
✅ Optimization Techniques
i. Linear Programming (LP)
ii. Integer Programming (IP)
Linear Programming (LP)
• Used to maximize or minimize a linear objective function, subject to linear
constraints.
• Suitable for problems like resource allocation, production planning, and budgeting.
Standard form:
Example: Maximize profit based on limited quantities of flour and sugar in a bakery.
Integer Programming (IP)
• A form of optimization where some or all decision variables must be integers.
• Useful when decisions involve yes/no, on/off, or quantity-based constraints.
• Applied in areas like employee scheduling, transportation planning, and capital
budgeting.
Example: Assigning delivery trucks (whole units) across multiple routes to
minimize cost.
✅ Simulation Modelling
1. Simulation involves creating a model to imitate a real-world
system and experimenting with it to evaluate outcomes under different
scenarios.
2. Helps understand uncertain or complex systems.
3. Often uses random variables to simulate real-life variability.
4. Common tools: Monte Carlo Simulation, Discrete Event Simulation.
Applications:
• Predicting delays in supply chain logistics.
• Modeling customer service queues in retail or banks.
• Evaluating financial risk under uncertain market conditions.
Steps in Prescriptive Analytics
Process
1. Define the Objective – e.g., maximize revenue, minimize cost.
2. Formulate the Model – includes objective function, decision variables, and
constraints.
3. Select a Technique – optimization or simulation based on the problem.
4. Solve the Model – using tools like Excel Solver, Python, R, etc.
5. Interpret Results – analyze the solution and test for feasibility.
6. Implement and Monitor – apply decisions and track performance
outcomes.
Tools Used in Prescriptive
Analytics
• Microsoft Excel (Solver add-in) – for basic LP and IP problems.
• Python libraries – e.g., PuLP, SciPy.optimize for optimization.
• R – packages like lpSolve, ROI.
• Simulation software – AnyLogic, Arena, Simul8, or Monte Carlo add-ins in
Excel.
Real-World Applications
Sector Prescriptive Analytics Use Cases
Route optimization, inventory control
Supply Chain
Healthcare Scheduling surgeries, allocating beds
Finance Portfolio optimization, credit risk modeling
Retail Price optimization, store layout design
Manufacturing Production planning, equipment maintenance
scheduling
Benefits and Challenges
✅ Benefits:
• Enhances strategic decision-making.
• Helps businesses respond quickly to change.
• Ensures optimal resource utilization.
⚠️Challenges:
• Requires high-quality data.
• Complex models need expertise and software tools.
• Implementation needs coordination across departments.
