Prescriptive Analytics goes beyond predicting what will happen — it recommends what actions to take for the best outcome. In this lesson, you’ll learn how optimization and decision models guide businesses toward smarter strategies and efficient resource use.
While descriptive analytics tells us what happened and predictive analytics forecasts what might happen, Prescriptive Analytics answers: What should we do next? It combines business rules, optimization algorithms, and simulations to recommend the best course of action.
Optimization is at the heart of prescriptive analytics. It involves selecting the best option from a set of possible choices while respecting given constraints. Common techniques include:
Decide the best inventory levels, delivery routes, and supplier mix to reduce cost and improve efficiency.
Optimize discounts and promotions to maximize sales without reducing profit margins.
Assign the right number of staff to shifts while minimizing overtime and meeting demand.
Allocate budgets or investments across projects for maximum ROI under given risk constraints.
from pulp import LpMaximize, LpProblem, LpVariable
# Problem: Maximize profit with limited resources
model = LpProblem(name="production-plan", sense=LpMaximize)
# Decision variables: number of product A and B
A = LpVariable("A", lowBound=0, cat="Integer")
B = LpVariable("B", lowBound=0, cat="Integer")
# Objective: profit = 20A + 30B
model += 20*A + 30*B
# Constraints
model += 2*A + 4*B <= 100 # labor hours
model += 3*A + 2*B <= 90 # raw materials
# Solve
model.solve()
print(f"Produce {A.value()} units of A and {B.value()} units of B")
This example finds the optimal production mix of two products given labor and material constraints.