Questions
4–6 questions per paper
Difficulty
Medium
Importance
High yield for HPCL/NTPC/ONGC
Overview
Operations Research and Business Statistics is a foundational module in PSU exams that focuses on mathematical modeling for decision-making and data interpretation. It evaluates your ability to optimize complex systems under constraints and apply statistical inferences to business processes, which is critical for industrial management roles.
Linear Programming (Simplex, Transportation, Assignment)
Linear Programming focuses on finding the optimal solution to a linear objective function subject to linear constraints. For PSU exams, mastering the Simplex method for optimality and the Vogel’s Approximation Method (VAM) for transportation is essential.
- Simplex method: Simplex criteria for optimality in maximization is zj - cj >= 0
- Transportation Problem: m + n - 1 number of basic feasible solutions
- Assignment Problem: Hungarian method complexity is O(n^3)
- Degeneracy occurs when number of occupied cells is less than m + n - 1
- Dual simplex is preferred when initial basic solution is infeasible but optimal
Queuing Theory
Queuing theory involves the mathematical study of waiting lines, often modeled as birth-death processes. Exam questions usually revolve around M/M/1 queuing models with Poisson arrivals and exponential service times.
- Utilization factor (rho) = lambda / mu
- Probability of zero units in system P0 = 1 - rho
- Average number of units in system Ls = lambda / (mu - lambda)
- Waiting time in queue Wq = rho / (mu - lambda)
- Little's Law: Ls = lambda * Ws
Probability Distributions & Hypothesis Testing
This section covers the behavior of random variables and the verification of population parameters based on sample statistics. You must be comfortable with standard distributions and test statistics like Z, t, and Chi-square.
- Poisson distribution variance equals mean (lambda)
- Normal distribution: 68.3% data within 1 standard deviation
- Null hypothesis (H0) is rejected if calculated value > critical value
- Type I error: Rejecting H0 when H0 is true
- Chi-square test is used for goodness of fit
Regression Analysis
Regression measures the strength and nature of the relationship between variables. PSU exams focus on the line of best fit and the calculation of correlation coefficients.
- Coefficient of determination R^2 = Explained variation / Total variation
- Correlation coefficient r lies between -1 and +1
- Regression coefficient b(yx) = r * (sigma y / sigma x)
- Standard error of estimate measures the accuracy of predictions
- Intercept 'a' calculation: a = mean(y) - b * mean(x)
Formula Sheet
Z = (x - mu) / sigma
Poisson: P(X=k) = (lambda^k * e^-lambda) / k!
Little's Law: L = lambda * W
Simplex optimality condition: zj - cj >= 0
R^2 = 1 - (SSE / SST)
Rho = lambda / mu
T-stat = (x_bar - mu) / (s / sqrt(n))
Exam Tip
Focus on identifying the 'bottleneck' in transportation problems and solving Queuing Theory using Little's Law, as these provide the fastest marks.
Common Mistakes
- Confusing the number of dummy rows/columns in balanced vs unbalanced assignment problems
- Forgetting to check the condition of lambda < mu in M/M/1 queuing systems, leading to negative results
- Misinterpreting one-tailed versus two-tailed tests during significance level calculation
More Revision Notes
Ready to test yourself?
Play topic-wise Operations Research & Business Statistics questions in Aspirant Arcade — gamified MCQ practice.
Download Free