Long Term Capacity Planning

Applied Capacity Expansion Optimization for Manufacturing Strategy

Project Date: 13 April 2024

---

Confidentiality Notice

The original company name and all identifying information in the datasets have been changed to protect confidentiality.
All data values, cost structures, and demand patterns have been anonymized or modified while remaining representative of a real world manufacturing capacity planning and Sales and Operations Planning scenario.
No real company data, proprietary information, or confidential business records are disclosed.

---

Overview

This project presents a long term capacity planning optimization model designed to support strategic manufacturing investment decisions.
The model evaluates multiple capacity expansion options over a multi year planning horizon and determines the optimal timing of investments required to meet forecasted demand while respecting annual budget constraints and construction lead times.

The project is structured as a professional analytics case study suitable for public GitHub portfolios, technical interviews, and applied roles in operations analytics, supply chain planning, and decision optimization.

---

Business Context

Manufacturing organizations must make capital intensive capacity decisions several years in advance. These decisions involve trade offs between demand growth, investment cost, operational feasibility, and financial constraints.

OptiManu Inc faces increasing demand volatility and must determine:

• When to invest in additional production capacity
• Which expansion options provide sufficient long term coverage
• Whether projected demand can be satisfied under annual budget limitations

---

Objectives

The optimization model is designed to:

• Satisfy forecasted demand in every planning year
• Respect annual budget constraints
• Select each expansion option at most once
• Incorporate construction lead times before capacity becomes operational
• Minimize total cost across the full planning horizon

---

Data Description

All datasets are anonymized and used strictly for demonstration and portfolio purposes.

Business_Planning_Data_2014_2024.csv

Year	Forecasted Demand	Operational Cost (USD)	Required Labor Hours	Required Machinery Hours	Average Wage (USD)	Workforce Size	Labor Market Tightness	Expected Total Revenue (USD)	Expected Raw Material Cost (USD)	Expected Compliance Cost (USD)	Expected Environmental Compliance Cost (USD)	Expected Labor Law Impact Cost (USD)	Expected Technology Investment Cost (USD)	Annual Budget (USD)
2014	50000	50000	1000	800	20	50	0.05	7000000	2500000	2000	1000	500	10000	6986000
2015	52000	51500	1100	850	22	55	0.055	7252000	2652000	2200	1100	550	10500	7237496
2016	54080	53000	1200	900	24	60	0.06	7516160	2812160	2400	1200	600	11000	7501128
2017	56243	54500	1300	950	26	65	0.065	7793029	2980879	2600	1300	650	11500	7777443
2018	58495	56000	1400	1000	28	70	0.07	8083480	3158730	2800	1400	700	12000	8067313
2019	60840	57500	1500	1050	30	75	0.075	9388200	3346200	3000	1500	750	12500	9369424
2020	63282	59000	1600	1100	32	80	0.08	10707892	3543792	3200	1600	800	13000	10686476
2021	65825	60500	1700	1150	34	85	0.085	9043275	3752025	3400	1700	850	13500	9025188
2022	68476	62000	1800	1200	36	90	0.09	9395408	3971608	3600	1800	900	14000	9376617
2023	71240	63500	1900	1250	38	95	0.095	9765160	4203160	3800	1900	950	14500	9745630

---

Expansion_Costs.csv

Proposed Expansion	Cost (USD)	Time to Build (years)	Additional Capacity (units)	Efficiency Gain
New Production Line	3000000	0.5	20000	0.10
Factory A (Small)	5000000	1	30000	0.15
Factory B (Medium)	8000000	2	50000	0.25
Factory C (Large)	12000000	3	80000	0.40

---

Methodology

Input data are validated for completeness, correct year indexing, and internal consistency prior to model execution.
A mixed integer optimization model is formulated using Pyomo with binary expansion selection variables, demand satisfaction constraints, annual budget constraints, and capacity availability adjusted for construction lead times.
The optimization model is defined in src/02_optimization_model.py and imported by the solver script.

---

Project Structure

optimanu_capacity_planning
├── README.md
├── data
│ ├── Business_Planning_Data_2014_2024.csv
│ └── Expansion_Costs.csv
├── src
│ ├── 01_data_preparation.py
│ ├── 02_optimization_model.py
│ └── 03_solve_and_report.py
├── results
│ ├── expansion_plan.csv
│ ├── annual_cost_breakdown.csv
│ └── summary_metrics.txt
└── figures

---

Reproducibility

`Environment requirements:

• Python 3.10 or later
• pandas
• pyomo
• GLPK solver (glpsol must be available in system PATH)

Execution logic:

• 01_data_preparation.py validates and prepares input data
• 02_optimization_model.py defines the optimization model
• 03_solve_and_report.py imports the model, solves it, and exports results

Execution order:

python src/01_data_preparation.py
python src/03_solve_and_report.py`

---

Analytical Value

The outputs support strategic analysis of optimal capacity investment timing, trade offs between capital expenditure and demand coverage, and long term budget feasibility.

---

Limitations

Demand and cost projections are treated as deterministic.
Capacity is modeled at an annual aggregate level.
Efficiency gains are not dynamically propagated across years.

---

Author

ABDOULAYE DIOP

Applied Data Science and Analytics