Capacity, Procurement and Supply Chain Operations

Introduction

A startup creates value only when it can reliably obtain inputs, transform them into products or services, and deliver them to customers. This operating system includes demand forecasting, capacity planning, scheduling, inventory control, procurement, supplier management, logistics and fulfilment. Weakness in any one area can produce stockouts, excess cash tied up in inventory, late delivery, quality failures or reputational damage.

For an entrepreneurial firm, operations decisions are especially important because resources are limited and demand is uncertain. The central managerial task is not simply to minimise cost. It is to balance cost, quality, speed, dependability, flexibility, resilience and ethical responsibility.

Learning Objectives

By the end of this chapter, you should be able to:

1. Explain the relationship between demand, capacity, inventory and supply-chain performance.
2. Prepare basic forecasts and identify their limitations.
3. Calculate and interpret capacity, utilisation, efficiency, reorder point and economic order quantity.
4. Develop practical schedules and fulfilment processes.
5. Evaluate suppliers and design purchasing controls.
6. Select appropriate logistics and inventory practices.
7. Assess technology, resilience, sustainability and ethical sourcing issues.

1. The Operations and Supply-Chain System

Operations management converts inputs into outputs. Inputs may include materials, labour, equipment, information, finance and supplier services. Outputs may be physical products, digital products or customer experiences. A supply chain links external and internal activities from upstream suppliers to the final customer.

A simple chain is:

supplier → purchasing → inbound transport → storage → production or service delivery → order fulfilment → customer → returns and feedback

The chain is not purely linear. Information about orders, forecasts, quality and customer preferences should flow in both directions. A supply chain therefore consists of three related flows:

A startup should map this system before making major commitments. A process map can reveal bottlenecks, duplicated handling, excessive lead times and single points of failure.

Figure: Material flow moves toward the customer while information should circulate throughout the chain.

2. Demand Forecasting

A forecast is a structured estimate of future demand, not a promise. Forecasts support purchasing, staffing, production, cash planning and delivery commitments. They should be prepared at the level at which decisions are made: for example, weekly demand for each product family rather than an unnecessarily precise daily estimate for every minor variant.

Qualitative and quantitative methods

Qualitative methods are useful when historical data are scarce. They include customer interviews, sales-force estimates, expert judgement, competitor analysis and the Delphi method. Quantitative methods use historical observations and may include moving averages, exponential smoothing, regression and seasonal analysis.

A simple moving average for the next period is:

Here, F_t+1 is the forecast for period t+1, D_t is actual demand in period t, and n is the number of periods averaged. All demand values must use the same unit and time interval. The method assumes that recent observations are reasonably representative and gives equal weight to each included period. It responds slowly to sudden changes.

Weighted moving average forecasting is:

The weights must satisfy w_i ≥ 0 and ∑w_i = 1. Larger weights can be assigned to more recent observations.

Exponential smoothing is:

where 0 < α ≤ 1 is the smoothing constant. A high α reacts quickly to change but may be unstable; a low value produces a smoother forecast. The method is most appropriate where level demand is present and strong trend or seasonality is absent unless extensions are used.

Forecast accuracy should be monitored rather than assumed. Common measures include mean absolute deviation:

and mean absolute percentage error:

MAD is expressed in demand units. MAPE is a percentage but becomes misleading when actual demand is zero or very close to zero. Forecast error should be reviewed by product, period and customer segment, not only as one aggregate number.

Worked example: moving average

A café sold 90, 110 and 100 lunch boxes over the last three days. A three-day moving-average forecast for day four is:

F₄ = (90 + 110 + 100) ÷ 3 = 100 lunch boxes.

If actual day-four demand is 120, the forecast error is 20 lunch boxes. The manager should investigate whether this reflects a promotion, a recurring weekday pattern or random variation before simply increasing future forecasts.

3. Capacity Planning and Scheduling

Capacity is the maximum output a resource can produce under stated conditions during a specified period. Design capacity is the theoretical maximum. Effective capacity is the realistic maximum after allowing for maintenance, breaks, changeovers, product mix and normal interruptions. Actual output is what was produced.

Two useful measures are:

The denominators differ. A business can have low utilisation because demand is weak, while still operating efficiently against its realistic capacity.

Capacity planning occurs at three levels:

• Strategic: premises, equipment, technology and outsourcing decisions.
• Tactical: staffing levels, supplier commitments, shifts and inventory policies.
• Operational: daily assignments, sequencing, dispatching and problem solving.

A bottleneck is a resource whose available capacity constrains total system output. Improving a non-bottleneck may not improve customer throughput. Managers should identify the bottleneck, protect it from interruptions, reduce unnecessary work around it and consider whether additional capacity is justified.

Scheduling converts plans into timed activities. Common rules include first come, first served; earliest due date; shortest processing time; and priority by customer or product criticality. No rule is universally best. For example, shortest processing time can reduce average waiting time but may delay large or urgent jobs. A practical schedule also includes setup time, material availability, labour skills, promised dates and contingency time.

For a simple workstation, required workload is:

If 240 units are required and each takes 0.25 labour-hours, required work is 60 labour-hours. If one employee provides 7.5 productive hours per day, at least 8 employee-days are required because 60 ÷ 7.5 = 8. This calculation assumes all hours are interchangeable and excludes additional setup or learning effects; real schedules should include them.

4. Inventory Management

Inventory may be raw material, work in progress, finished goods, maintenance supplies or goods in transit. Inventory creates benefits—availability, economies of scale and protection against uncertainty—but also incurs purchase, ordering, holding, damage, obsolescence and financing costs.

Reorder point and safety stock

A basic reorder point is:

where d is average demand per time unit, L is supplier lead time in the same time unit, and SS is safety stock. The term dL represents expected demand during lead time. The formula assumes reasonably stable average demand and lead time; it is not a guarantee against extreme variation.

If daily demand is 40 units, lead time is 5 days and safety stock is 30 units, then ROP = (40 × 5) + 30 = 230 units. A replenishment order is triggered when inventory position reaches approximately 230 units. Inventory position should normally include on-hand inventory plus on-order inventory minus backorders, not merely the physical quantity on the shelf.

Safety stock should reflect the desired service level, demand variability and lead-time variability. A more statistical model for constant lead time and approximately independent demand is:

where z is the standard-normal service factor, σ_d is the standard deviation of demand per period and L is lead time in periods. This model assumes demand is approximately normally distributed, lead time is fixed, and the selected service factor is appropriate. If both demand and lead time vary, a different model is required; the simple equation should not be applied blindly.

Economic order quantity

The basic economic order quantity is:

where D is annual demand in units, S is ordering or setup cost per order, and H is annual holding cost per unit. The classical model assumes known and constant demand, constant lead time, instantaneous replenishment, no quantity discounts, no shortages and stable costs. It is a benchmark, not an automatic purchase instruction.

For annual demand of 12,000 units, ordering cost of 25 currency units per order and annual holding cost of 2 currency units per unit:

EOQ = √[(2 × 12,000 × 25) ÷ 2] = √300,000 ≈ 548 units.

The startup would compare this result with supplier minimum order quantities, storage limits, cash availability and expiry risk.

Inventory classification and control

ABC analysis ranks items by annual usage value, calculated as annual quantity used × unit cost. A-items receive close managerial control; B-items receive moderate control; C-items may use simpler systems. Value is not the only criterion. A low-cost safety component may be operationally critical, while an expensive item may be easily substituted.

Useful inventory indicators include inventory turnover:

and days of inventory on hand:

These measures require consistent valuation and time periods. High turnover is not automatically good if it causes stockouts; low turnover may be justified for strategic or seasonal inventory.

5. Procurement and Supplier Management

Procurement is the organised process of identifying needs, specifying requirements, selecting suppliers, agreeing terms, ordering, receiving, inspecting and evaluating performance. Purchasing is one transaction within that broader process.

A sound procurement cycle is:

1. Define the need and specification.
2. Confirm budget, approval authority and quantity.
3. Identify and prequalify potential suppliers.
4. Request quotations or proposals.
5. Compare total cost and non-price factors.
6. Negotiate and issue a purchase order or contract.
7. Receive, inspect and record goods or services.
8. Match the purchase order, receipt and invoice.
9. Review supplier performance and renew, improve or replace the relationship.

The total cost of ownership may include purchase price, freight, duties, inspection, installation, maintenance, energy, defects, returns, switching costs and disposal. Selecting the lowest quoted price can therefore produce the highest real cost.

Supplier evaluation criteria commonly include:

Contracts should define specifications, quantities, prices, delivery terms, acceptance criteria, warranty, intellectual property, confidentiality, payment, liability, termination, dispute resolution and compliance obligations. Startups should avoid informal commitments that create uncontrolled dependency.

Segregation of duties is a basic control: where practical, different people should approve a purchase, receive goods and authorise payment. A three-way match between purchase order, receiving record and invoice helps prevent duplicate or unauthorised payment.

6. Logistics and Fulfilment

Logistics coordinates movement and storage. Inbound logistics brings inputs to the operation; internal logistics moves materials within it; outbound logistics delivers customer orders. Fulfilment includes order capture, payment confirmation, picking, packing, dispatch, tracking, delivery and returns.

A fulfilment design should specify:

• order cut-off times and promised service levels;
• inventory locations and stock allocation rules;
• picking method and confirmation controls;
• packaging, labelling and product protection;
• carrier selection and shipment tracking;
• failed-delivery and returns procedures;
• customer communication and escalation.

The choice between in-house fulfilment, a third-party logistics provider and a hybrid model depends on volume, geographic reach, product complexity, capital and strategic control. Outsourcing can provide expertise and variable cost, but introduces coordination, data-sharing and dependency risks.

Important performance measures include on-time-in-full delivery, order-cycle time, order accuracy, fill rate, damage rate, return rate, logistics cost per order and customer complaints. Measures should be defined precisely. For example, on-time-in-full requires both delivery by the promised date and the complete quantity requested.

7. Resilience, Sustainability and Ethical Sourcing

Efficiency-oriented supply chains may minimise inventory and suppliers, but excessive lean operation can increase vulnerability. Resilience is the ability to prepare for, respond to and recover from disruption. Sources of disruption include supplier insolvency, transport interruption, cyberattack, extreme weather, regulatory change, quality failure and geopolitical conflict.

Practical resilience measures include supplier diversification, approved substitutes, safety stock for critical items, dual sourcing, documented workarounds, insurance, business-continuity plans and regular scenario testing. These measures have costs, so they should be prioritised according to probability, impact and recovery time.

Ethical sourcing requires attention to forced labour, child labour, unsafe workplaces, discrimination, excessive working hours, corruption, misleading environmental claims and harmful disposal. A supplier code of conduct is useful only when supported by due diligence, contractual requirements, evidence and corrective action. Audits alone can miss hidden subcontracting; meaningful assessment may require worker interviews, traceability and independent verification.

Sustainability decisions include material selection, packaging reduction, energy use, transport mode, repairability, reuse, recycling and reverse logistics. A startup should distinguish measurable improvements from unsupported green claims. Environmental and social objectives should be included in procurement specifications and supplier reviews rather than treated as marketing language.

8. Digital Systems and Artificial Intelligence

Enterprise resource planning, inventory-management systems, barcode or RFID scanning, electronic purchasing, transport-management systems and customer-order platforms can improve visibility. However, technology does not correct inaccurate master data, poor processes or unclear accountability.

Artificial intelligence can support demand sensing, anomaly detection, route planning, supplier-risk monitoring and customer-service automation. Human managers remain responsible for defining objectives, validating data, checking bias, protecting confidential information and approving consequential decisions. Forecast models can amplify historical bias or fail during unprecedented events. A responsible system therefore records data sources, assumptions, confidence limits and override decisions.

A small startup should begin with reliable item codes, units of measure, supplier records, lead times, reorder rules and transaction discipline before purchasing complex software. Integration should be evaluated by total cost, security, scalability, interoperability and user adoption.

9. Illustrative Case: GreenCrate Foods

Background

GreenCrate Foods is an illustrative meal-kit startup selling chilled weekly boxes. It purchases ingredients from local farms, packs boxes in a rented facility and uses a courier network for delivery.

Problem

Demand grew from 400 to 900 boxes per week. The company experienced missing ingredients, emergency purchases, 12% late deliveries and excessive spoilage. Managers placed orders by message and spreadsheet, with no agreed supplier lead times or inventory records.

Analysis

A process review found three major causes. First, forecasts used optimistic sales targets rather than confirmed orders. Second, the packing table was the bottleneck because ingredient kits arrived unsorted. Third, suppliers were selected mainly on price, although a cheaper supplier had unreliable delivery. Chilled ingredients also had short shelf lives, making large safety stocks inappropriate.

Approach and solution

GreenCrate introduced a weekly demand process based on paid orders plus a documented estimate for likely late orders. It classified ingredients by value and criticality, established supplier lead times and created reorder points for stable items. For highly perishable items, it used confirmed-order purchasing and daily receiving checks.

The firm measured packing capacity by boxes per productive hour, redesigned the workstation, and scheduled ingredients in packing sequence. It introduced purchase orders, receiving records and a three-way invoice match. A supplier scorecard tracked quality, delivery reliability, responsiveness and ethical requirements. One critical ingredient received an approved substitute and a second supplier.

For fulfilment, orders were cut off 48 hours before dispatch, labelled by delivery route and scanned at packing. Customers received tracking messages and a clear returns policy.

Outcome

After eight weeks, late deliveries fell substantially, spoilage decreased and emergency purchases became less frequent. The company did not eliminate uncertainty; instead, it made assumptions visible, assigned responsibility and used data to revise decisions. The main lesson is that growth requires operating discipline before additional sales volume is pursued.

10. Implementation Guidelines and Limitations

A practical implementation sequence is:

1. Map the end-to-end process and identify customer requirements.
2. Establish accurate product, supplier, lead-time and inventory records.
3. Measure demand, capacity, quality, delivery and cash effects.
4. Identify bottlenecks and critical supply dependencies.
5. Introduce simple forecasting, purchasing and replenishment rules.
6. Document approval, receiving, payment and exception procedures.
7. Review supplier ethics, resilience and total cost.
8. Automate repetitive controls only after the process is stable.
9. Conduct monthly performance reviews and scenario tests.

Quantitative models are simplifications. Forecasts can fail when markets change; EOQ can be unsuitable for variable or perishable demand; reorder points depend on credible lead-time data; utilisation can encourage harmful overloading; and performance metrics can be manipulated if definitions are unclear. Managerial judgement, customer communication and ethical evaluation remain essential.

Chapter Summary

Capacity and supply-chain operations connect demand to reliable customer delivery. Forecasts provide estimates that guide purchasing, staffing and production, but must be monitored for error and revised when conditions change. Capacity planning distinguishes design capacity, effective capacity and actual output, while bottleneck analysis directs improvement toward the resource that limits throughput.

Inventory provides availability and protection but consumes cash and creates holding, obsolescence and quality risks. Reorder point, safety stock, EOQ, ABC analysis and turnover are useful tools when their assumptions are understood. Procurement should evaluate total cost, quality, delivery, flexibility, resilience and ethical conduct, not merely quoted price. Logistics and fulfilment require defined service promises, accurate picking, reliable transport, returns control and meaningful performance measures.

Resilient and ethical supply chains use diversification, continuity planning, traceability and responsible sourcing. Digital systems and AI can improve visibility and decision support, but only when data quality, security, accountability and human oversight are present. Effective entrepreneurial operations balance efficiency with flexibility, customer value, worker welfare, environmental responsibility and the capacity to grow.

Key Terms

Capacity: Maximum output possible under stated conditions and time.

Bottleneck: Resource that limits the throughput of the whole system.

Forecast: Structured estimate of future demand.

Lead time: Time between placing an order and receiving it or completing a process.

Safety stock: Additional inventory held against uncertainty.

Reorder point: Inventory position at which replenishment is triggered.

Economic order quantity: Benchmark order size that balances ordering and holding costs under stated assumptions.

Inventory position: On-hand inventory plus on-order inventory minus backorders.

ABC analysis: Classification of inventory according to usage value or managerial importance.

Procurement: End-to-end process of acquiring goods and services.

Purchase order: Authorised document specifying a purchase agreement.

Total cost of ownership: All relevant costs over the life of an acquisition.

Supplier scorecard: Structured evaluation of supplier performance.

Logistics: Planning and control of movement and storage.

Fulfilment: Activities from order receipt through delivery and returns.

Service level: Agreed or measured ability to meet customer demand or promises.

Resilience: Ability to withstand, respond to and recover from disruption.

Ethical sourcing: Acquisition that respects labour, human-rights, environmental and integrity standards.

Three-way match: Comparison of purchase order, receiving record and supplier invoice.

Review Questions

Short Answer Questions

1. Distinguish design capacity from effective capacity.
2. Why is a forecast not the same as a promise?
3. State two benefits and two costs of holding inventory.
4. What information is compared in a three-way match?
5. Identify three criteria, other than price, for evaluating suppliers.

Long Answer Questions

1. Explain how demand forecasting, capacity planning and scheduling should be coordinated in a growing startup.
2. Describe the assumptions and managerial uses of the basic EOQ model.
3. Explain how a startup can balance efficiency with supply-chain resilience.
4. Discuss the role of procurement controls in protecting cash, quality and ethical standards.
5. Evaluate the operational requirements of a reliable fulfilment system.

Analytical / Application Questions

1. A business has daily demand of 60 units, a lead time of 4 days and safety stock of 50 units. Calculate the reorder point and explain what data could make this calculation unreliable.
2. A firm’s design capacity is 1,000 units per month, effective capacity is 800 units and actual output is 720 units. Calculate utilisation and efficiency, then interpret the difference.
3. Design a procurement and fulfilment improvement plan for a startup that has rapid sales growth, frequent stockouts, one dominant supplier and increasing customer complaints about late deliveries.

Multiple Choice Questions with Answers

Question 1

Which statement best describes a forecast?

A. A guaranteed quantity that customers will purchase

B. A structured estimate of future demand

C. A legal commitment made to a supplier

D. The maximum output of a production system

Correct Answer: B. A structured estimate of future demand

Question 2

A firm has actual output of 720 units and design capacity of 1,000 units. What is utilisation?

A. 28%

B. 72%

C. 80%

D. 128%

Correct Answer: B. 72%

Question 3

If average daily demand is 40 units, lead time is 5 days and safety stock is 30 units, what is the reorder point?

A. 170 units

B. 200 units

C. 230 units

D. 250 units

Correct Answer: C. 230 units

Question 4

Which condition is an assumption of the basic EOQ model?

A. Demand is known and constant

B. Lead time changes randomly every day

C. Shortages are required

D. Holding cost is zero

Correct Answer: A. Demand is known and constant

Question 5

Which measure compares actual output with effective capacity?

A. Utilisation

B. Efficiency

C. Inventory turnover

D. Fill rate

Correct Answer: B. Efficiency

Question 6

What is the main purpose of safety stock?

A. To eliminate all purchasing activity

B. To protect against demand or lead-time uncertainty

C. To increase product obsolescence

D. To replace supplier evaluation

Correct Answer: B. To protect against demand or lead-time uncertainty

Question 7

Which item is normally part of total cost of ownership?

A. Only the quoted unit price

B. Only the supplier’s advertising cost

C. Freight, maintenance, defects and disposal costs

D. The customer’s personal preferences only

Correct Answer: C. Freight, maintenance, defects and disposal costs

Question 8

What does a three-way match compare?

A. Forecast, budget and marketing plan

B. Purchase order, receiving record and invoice

C. Supplier, competitor and customer

D. Capacity, utilisation and efficiency

Correct Answer: B. Purchase order, receiving record and invoice

Question 9

Which action most directly improves supply-chain resilience?

A. Relying on one untested supplier

B. Removing all contingency stock

C. Identifying approved substitutes and alternative suppliers

D. Ignoring supplier lead-time variability

Correct Answer: C. Identifying approved substitutes and alternative suppliers

Question 10

Why should AI-supported supply decisions retain human oversight?

A. AI systems never use data

B. Models can fail under unusual conditions and reproduce data bias

C. Human oversight always lowers accuracy

D. AI cannot perform arithmetic

Correct Answer: B. Models can fail under unusual conditions and reproduce data bias