QUANTITATIVE TECHNIQUES
CHAPTER 6: LINEAR PROGRAMMING
Linear Programming (LP) is a mathematical technique used to allocate limited
resources such as time, money, labour, or raw materials in the most efficient way
possible. In logistics and supply chain management, LP is widely used to minimise
costs, maximise profits, and ensure optimal allocation of resources.
For South African logistics contexts, LP supports decisions such as:
- Allocating trucks to routes between Johannesburg, Durban, and Cape Town.
- Balancing diesel vs. electric vehicles in a logistics fleet.
- Scheduling production of bread and scones in township bakeries with limited flour
and labour.
- Assigning warehouse space for multiple clients under capacity constraints.
LEARNING OUTCOMES
1. Define linear programming and explain its role in logistics and supply chain
management.
2. Formulate objective functions and constraints for real-world problems.
3. Graph linear inequalities and identify feasible regions.
4. Apply the corner-point method to determine optimal solutions.
5. Solve LP problems using Microsoft Excel 365 Solver.
6. Use OpenAI to critique, refine, and interpret LP models.
7. Relate LP to the Sustainable Development Goals (SDGs).
KEY CONCEPTS
1. Decision Variables – What we control (e.g., number of diesel vans vs. electric
vans).
2. Objective Function – What we want to optimise (maximise profit or
minimise cost).
3. Constraints – Resource limitations written as inequalities (≤, ≥).
4. Feasible Region – All possible solutions that satisfy the constraints.
5. Optimal Solution – The best point in the feasible region, usually at a
corner.
1
Tip: To visualise LP problems, we graph inequalities on a coordinate system.
The overlapping shaded area is the feasible region, and the optimal solution lies
on one of its corners.
STEP-BY-STEP LINEAR PROGRAMMING EXAMPLES
Example 1: Township Bakery – Bread and Scones
A bakery produces bread (B) and scones (S). Profit = R40 per bread batch and R30
per scone batch.
Constraints:
- Flour: 2𝐵 + 𝑆 ≤ 100
- Time: 𝐵 + 𝑆 ≤ 80
- 𝐵, 𝑆 ≥ 0
Objective Function:
Maximise 𝑍 = 40𝐵 + 30𝑆
Feasible region corners: (0,0), (0,80), (20,60), (50,0)
2
• Evaluating profits:
- (0,80): Z = R2,400
- (20,60): Z = R2,600 (optimal)
- (50,0): Z = R2,000
Optimal Solution: 20 batches of bread, 60 batches of scones → Profit
R2,600.
Example 2: Furniture Company – Tables and Chairs
A furniture workshop produces tables (T) and chairs (C). Profit = R120 per table,
R80 per chair.
Constraints:
- Assembly: 3𝑇 + 2𝐶 ≤ 200
- Finishing: 2𝑇 + 2𝐶 ≤ 180
- Inspection: 𝑇 + 𝐶 ≤ 40
- 𝑇, 𝐶 ≥ 0
Objective Function:
Maximise 𝑍 = 120𝑇 + 80𝐶
3
Feasible region corners: (0,0), (0,40), (20,20), (60,0)
Evaluating profits:
- (20,20): Z = R4,000 (optimal)
Optimal Solution: 20 tables, 20 chairs → Profit R4,000.
Example 3: Transport Allocation – Diesel vs. Electric Vans
A logistics company assigns diesel vans (X1) and electric vans (X2) to routes.
Constraints:
- Availability: X1 + X2 ≤ 100
- Diesel minimum: X1 ≥ 20
- Electric minimum: X2 ≥ 10
Objective Function:
Minimise 𝑍 = 50𝑋1 + 80𝑋2
Feasible region corners: (20,10), (20,80), (90,10)
Evaluating costs:
4
- (20,10): Z = R1,800 (optimal)
Optimal Solution: 20 diesel vans, 10 electric vans → Cost R1,800.
FORECASTING WITH MICROSOFT EXCEL 365 SOLVER
Excel provides a powerful Solver tool to automate LP solutions:
1. Enter the objective function in a cell (e.g., Profit = 40B + 30S).
2. Enter decision variables (B, S) in cells.
3. Define constraints (e.g., flour, time).
4. Use Solver → Set Objective → By Changing Variables → Add Constraints →
Solve.
OPENAI INTEGRATION
OpenAI can help you:
- Formulate models: 'Write the LP constraints for a bakery producing bread and
scones.'
- Interpret results: 'Explain why the corner-point method finds the optimal
solution.'
- Test scenarios: 'What happens if diesel costs increase to R60 per van?'
CASE STUDY CONNECTIONS
1. Retail Distribution (South Africa): Allocate trucks between Johannesburg,
Durban, and Cape Town to minimise transport costs.
2. Mining: Optimise allocation of machinery and labour in a gold mine.
3. Education: Plan survey sampling across provinces while minimising costs (similar
to UJ research projects).
SUSTAINABLE DEVELOPMENT GOALS (SDGs)
- SDG 8: Decent Work and Economic Growth → Efficient resource use boosts
productivity.
- SDG 9: Industry, Innovation, and Infrastructure → LP supports cost-efficient
infrastructure planning.
- SDG 13: Climate Action → Balancing diesel vs. electric vehicles reduces emissions.
5
PRACTICAL ACTIVITIES
– Formulate an LP model for allocating warehouse space in Johannesburg. Solve in
Excel Solver.
– Re-create the Bakery LP graph using the Math is Fun inequality plotting guide.
– Use OpenAI to critique your LP formulation and suggest alternative constraints.
– Solve a 'diet problem' in Excel: minimise cost of meals while meeting nutrition
constraints.
REFLECTION QUESTIONS
– Why are corner points critical in LP solutions?
– How does LP support decision-making in logistics?
– What are the advantages of Excel Solver vs. manual graphical methods?
– How can LP contribute to achieving SDG 13 (Climate Action)?
SOURCES
Render, B., Stair, R. M., Hanna, M. E., & Hale, T. S. (2018). Quantitative analysis for
management (13th ed.). Pearson.
Louw, A. (2024). Linear Programming Worksheets (Unpublished class material).
University of Johannesburg.
University of Johannesburg. (2025). Transport Supply Management Calculations –
Course Update Document.
Math is Fun. (n.d.). Graphing linear inequalities.
[Link]
Case data adapted to South African logistics contexts: township bakeries, Durban
Port, Gautrain, and transport fleets.