Capacity & Location Planning: Models, Trade-offs & Strategic Decisions

TAKEAWAYS:

• Capacity planning balances utilization rate (70–90% ideal) with costs (underutilization vs. overinvestment) using break-even analysis and learning curves.
• Location decisions rely on factor-rating models (qualitative) and center-of-gravity methods (quantitative), but political risks (e.g., Nepal’s trade blockades) often override math.
• Aggregate planning (e.g., Daraz’s holiday hiring) combines inventory, workforce, and subcontracting to match demand, using linear programming for optimization.
• Capacity cushion (extra capacity for demand spikes) costs money but prevents lost sales—Nepal’s NTC uses this for monsoon-season network upgrades.
• Location trade-offs (e.g., Kathmandu vs. Pokhara) pit land costs against labor skills and supply chain proximity (e.g., Chaudhary Group’s warehouses near highways).
• Sustainability is now a location factor: Google’s data centers use cool climates (e.g., Finland) to cut energy costs, while Nepal’s hydropower attracts manufacturers like Himalayan Java.

1. Capacity Planning: Matching Supply to Demand

Capacity is the maximum output a system (machine, worker, factory) can produce per unit time. Poor capacity planning leads to:

• Underutilization (wasted resources, e.g., idle Daraz warehouses in off-season).
• Overutilization (burnout, delays, e.g., Ncell’s network crashes during festivals).

Key Concepts

How to Plan Capacity

1. Forecast Demand: Use time-series analysis (e.g., Pathao’s ride demand peaks at 8 PM).
2. Determine Capacity Requirements: Calculate needed machines/workers.
3. Choose Expansion Strategy:
   • Lead Strategy: Expand before demand rises (high risk, low stockouts; e.g., Daraz adding warehouses pre-Dussehra).
   • Lag Strategy: Expand after demand proves steady (low risk, high stockouts; e.g., Khalti scaling servers post-lockdown).
   • Match Strategy: Expand with demand (balanced; e.g., Nepal Rastra Bank adjusting cash supply).

flowchart TD
    A["Demand Forecast"] --> B["Current Capacity"]
    B --> C{"Capacity Shortage?"}
    C -->|"Yes"| D["Expand: Lead/Lag/Match"]
    C -->|"No"| E["Monitor"]
    D --> F["Re-evaluate"]
    F --> C

Worked Example: Kathmandu Traffic Routes

Problem: Kathmandu’s Ring Road has a design capacity of 100,000 vehicles/day but sees 150,000 on weekends. Utilization = 150% → bottlenecks. Solutions:

1. Add Lanes (increase effective capacity; cost: ₹500M).
2. Smart Traffic Lights (improve flow; cost: ₹50M).
3. Public Transport Incentives (reduce demand; cost: ₹200M). Decision: Use break-even analysis to pick the cheapest option that avoids ₹10M/day in lost productivity.

Break-Even Formula: $$\text{Break-Even Point (in vehicles)}=\frac{\text{Fixed Cost}}{\text{Cost per Vehicle}-\text{Variable Cost}}$$ For smart lights:

• Fixed Cost = ₹50M
• Cost per Vehicle = ₹0.10 (reduced travel time)
• Variable Cost = ₹0.05 (maintenance) $$\text{Break-Even}=\frac{50,000,000}{0.10-0.05}=1,000,000\text{ vehicles}$$ Since Kathmandu sees 150,000/day, smart lights pay off in 7 years.

2. Location Planning: Where to Operate?

Location affects costs, risks, and competitiveness. Key factors:

• Proximity to Markets (e.g., Daraz warehouses near cities).
• Proximity to Suppliers (e.g., Himalayan Java near coffee farms).
• Labor Costs/Skills (e.g., Bhatbhateni for textile workers).
• Infrastructure (e.g., NTC towers in hilly areas vs. flatlands).
• Government Incentives (e.g., SEEPZ in India for tax breaks).
• Climate/Risks (e.g., flood-prone areas for hydropower like Kulekhani).

Location Decision Models

mindmap
  root((Location Planning))
    Factor-Rating
      Step 1: List Factors (e.g., Cost, Labor, Risk)
      Step 2: Assign Weights (e.g., Cost=40%, Labor=30%)
      Step 3: Score Locations (1-5)
      Step 4: Calculate Weighted Score
    Center-of-Gravity
      Formula: X = Σ(x_i × Q_i)/ΣQ_i, Y = Σ(y_i × Q_i)/ΣQ_i
      Q_i = Demand at point i
    Transportation Model
      Linear Programming to minimize cost
      Constraints: Supply ≤ Demand

Worked Example: Nabil Bank’s New Branch

Problem: Open a branch in Pokhara vs. Bhaktapur. Factors:

• Pokhara: (5×30 + 3×20 + 4×25 + 3×15 + 2×10) = 4.15
• Bhaktapur: (3×30 + 5×20 + 2×25 + 5×15 + 5×10) = 3.55 Decision: Pokhara wins (higher score), but Bhaktapur’s tax break might offset the difference.

3. Aggregate Planning: Balancing Demand and Capacity

Aggregate planning matches supply and demand over 2–18 months using:

• Inventory (hold stock; e.g., Daraz stocking Diwali gifts early).
• Workforce (hire/fire; e.g., NTC temporary workers during Dasain).
• Overtime/Idle Time (e.g., Himalayan Java roasting extra beans in peak season).
• Subcontracting (e.g., Nepal Rastra Bank outsourcing IT support).

Strategies

Graph: Aggregate Planning Options

Worked Example: Daraz’s Holiday Hiring

Problem: Daraz expects 30% more orders in Dussehra (1 month). Options:

1. Hire 30% more staff (cost: ₹5M for salaries).
2. Use overtime (cost: ₹3M, but 50% overtime pay).
3. Increase inventory (cost: ₹4M for extra stock). Decision: Mixed strategy (20% hiring + 10% overtime) to avoid ₹2M/day in lost sales from stockouts.

4. Real-World Applications in Nepal

Case Study 1: NTC’s Network Capacity Planning

• Challenge: Nepal’s monsoon rains damage towers, reducing capacity by 30%.
• Solution:
  • Capacity cushion: Adds 20% extra fiber before monsoon.
  • Predictive maintenance: Uses AI to predict tower failures (partnership with Ncell).
• Result: 99.8% uptime during rains (vs. 95% without planning).

Case Study 2: Chaudhary Group’s Warehouse Location

• Problem: Daraz needed warehouses near Kathmandu, Pokhara, and Biratnagar.
• Method: Center-of-gravity model to minimize shipping costs.
• Outcome:
  • Kathmandu: High demand (weight = 50%).
  • Pokhara: Medium demand (30%).
  • Biratnagar: Low demand (20%).
• Result: Warehouses built near highways (e.g., Pradakshin Marg) to cut transport costs by 15%.

Case Study 3: Himalayan Java’s Climate-Based Location

• Challenge: Coffee beans need consistent humidity.
• Solution: Factories in Dhankuta (cool, high-altitude) vs. hot Terai.
• Result: 20% higher yield due to ideal drying conditions.

In the Real World

1. eSewa’s Server Capacity:

   • Idea: Load balancing (distributing traffic across servers).
   • How: During Dashain, eSewa’s servers in Kathmandu, Pokhara, and India share the load to avoid crashes.
   • Impact: 99.9% uptime even with 5× normal transactions.

2. Pathao’s Driver Scheduling:

   • Idea: Chase demand strategy (adjusting driver supply).
   • How: Pathao hires 30% more drivers during Tihar and lays them off post-festival.
   • Impact: 30% lower wait times during peak hours.

3. Nepal Rastra Bank’s Cash Supply:

   • Idea: Aggregate planning (balancing cash inventory).
   • How: NRB increases cash in ATMs by 40% before Dashain and reduces it afterward.
   • Impact: Avoids ₹500M/year in lost productivity from cash shortages.

Exam Tip

1. Numerical Problems (30% of marks):

   • Break-even analysis: Always show the formula and units (e.g., "₹ per unit").
   • Center-of-gravity: Plot points on graph paper and label axes (e.g., "Distance in km").
   • Aggregate planning: Compare costs of strategies in a table.

2. Case Studies (25% of marks):

   • Structure:
     1. Problem (e.g., "NTC’s monsoon outages").
     2. Factors considered (e.g., "cost, risk, technology").
     3. Solution (e.g., "20% capacity cushion + AI").
     4. Outcome (e.g., "99.8% uptime").
   • Example: For Daraz’s warehouse, mention center-of-gravity and highway proximity.

3. Short Answers (20% of marks):

   • Define:
     • Capacity cushion: "Extra capacity to handle demand spikes."
     • Factor-rating: "Qualitative method to score locations."
   • Compare:

     Lead Strategy	Lag Strategy
     High risk	Low risk
     Low stockouts	High stockouts
     Example: Daraz	Example: Khalti

4. Diagrams (15% of marks):

   • Draw:
     • Break-even graph (plot cost vs. volume).
     • Center-of-gravity (show demand points and optimal location).
     • Aggregate planning (show demand curve vs. production levels).
   • Label: Always include axes, lines, and key points.

5. Common Pitfalls:

   • Ignoring qualitative factors (e.g., "political risk" in Nepal).
   • Forgetting units in calculations (e.g., "₹ per hour" vs. just "₹").
   • Overlooking sustainability (e.g., "Google’s cool climates" for data centers).

flowchart LR
    A["Capacity Planning"] --> B["Forecast Demand"]
    A --> C["Determine Capacity"]
    A --> D["Choose Strategy\n(Lead/Lag/Match)"]
    D --> E["Re-evaluate"]
    E --> A
    B --> F["Time-Series\nRegression"]
    C --> G["Utilization Rate\n(70-90% ideal)"]
    D --> H["Break-Even Analysis"]
    H --> I["Cost Comparison"]