Operations Research

Why it matters

Operations Research (OR) is crucial in industrial engineering as it provides systematic and quantitative methods to optimize complex processes, improve decision-making, and enhance efficiency in manufacturing and service operations. It helps in resource allocation, scheduling, and minimizing costs, which are vital for competitive advantage in industries.

Key ideas

• Optimization: The core of OR, focusing on finding the best solution from a set of feasible solutions.
• Linear Programming (LP): A mathematical technique for optimizing a linear objective function, subject to linear equality and inequality constraints.
• Queuing Theory: Analyzes waiting lines or queues to improve service efficiency.
• Simulation: Uses models to replicate the operation of a system to study its behavior under different scenarios.
• Decision Analysis: Involves making decisions under uncertainty using tools like decision trees and utility theory.
• Network Models: Used for optimizing logistics and supply chain operations, including shortest path, maximum flow, and minimum cost flow problems.

Formulas

• Z = c1*x1 + c2*x2 + ... + cn*xn

  • Z: Objective function value (units depend on the model, e.g. rupees)
  • c1, c2, ..., cn: Coefficients of the objective function (unit depends on context)
  • x1, x2, ..., xn: Decision variables (unit depends on context)

• Mean service time = 1 / μ

  • Arrival rate λ and service rate μ are independent model inputs; λ is not the reciprocal of μ
  • μ: Service rate (customers per unit time)

Worked example

Problem: A factory produces two products, A and B. The profit per unit of A is ₹40 and B is ₹30. The production constraints are:

1. Maximum of 60 hours of labor available.
2. Product A requires 2 hours per unit, and B requires 1 hour per unit.
3. Maximum of 50 units of raw material available.
4. Product A requires 1 unit of raw material, and B requires 1 unit per unit.

Find the optimal production mix to maximize profit.

Solution:

1. Define decision variables:
   • Let x be the number of units of A produced.
   • Let y be the number of units of B produced.
2. Formulate the objective function:
   • Maximize Z = 40x + 30y
3. Formulate the constraints:
   • 2x + y ≤ 60 (labor constraint)
   • x + y ≤ 50 (material constraint)
   • x ≥ 0, y ≥ 0 (non-negativity constraint)
4. Solve using graphical or simplex method.
   • Feasible corner points are (0,0), (30,0), (10,40) and (0,50), with profits 0, 1200, 1600 and 1500 respectively. Optimal solution: x = 10, y = 40
   • Maximum profit: Z = 40(10) + 30(40) = 400 + 1200 = 1600

Final Answer: ₹1600, achieved with 10 units of A and 40 of B. The earlier candidate (20,30) would require 70 labor-hours and is infeasible.

Common mistakes

• Misidentifying decision variables and constraints.
• Incorrectly setting up the objective function or constraints.
• Ignoring non-negativity constraints.
• Failing to check the feasibility of the solution.

For GATE ME

Questions often involve formulating and solving linear programming problems, analyzing queuing systems, and applying network models. Practice setting up and solving LP problems, understanding queuing theory concepts, and using network optimization techniques.

Quick check

1. What is the primary goal of operations research?
2. Name a common method used in operations research for optimizing linear problems.
3. What does queuing theory analyze?

Answers: 1. Optimization of processes and decision-making. 2. Linear Programming. 3. Waiting lines or queues.