Process Planning, Production Analytics, and Economics

Introduction and Context

A manufactured component is not produced by isolated machine operations. It is produced through a coordinated sequence in which design requirements, material behavior, machine capability, tooling, inspection, labor, energy, and delivery commitments interact. Process planning converts a product drawing and specification into an economical, safe, and controllable manufacturing route. Production analytics evaluates whether that route is performing as intended. Manufacturing economics determines whether the proposed method creates sufficient value compared with alternatives.

These activities are strategically important because a poor process plan can lock an organization into excessive setup time, unnecessary movement, low machine utilization, high scrap, or unreliable delivery. Conversely, a sound plan balances quality, cost, flexibility, capacity, and risk rather than minimizing any one variable in isolation.

Learning Objectives

After studying this chapter, the student should be able to:

1. Interpret product requirements and sequence machining operations logically.
2. Select machines, workholding methods, tools, cutting conditions, and inspection stages.
3. Estimate machining time, labor cost, tooling cost, and total manufacturing cost.
4. Analyze capacity, utilization, throughput, bottlenecks, and overall equipment effectiveness.
5. Evaluate trade-offs among quality, productivity, flexibility, automation, and risk.
6. Make reasoned tooling, automation, and make-or-buy decisions.
7. Use production data ethically and critically to support improvement decisions.

1. Process Planning as a Transformation System

A process plan specifies how a part will be transformed from raw material into an accepted product. It normally contains:

• material, blank size, and preparation method;
• datum scheme and workholding arrangement;
• operation sequence and machine assignment;
• cutting tools, tool materials, and tool geometry;
• cutting speed, feed, depth of cut, and coolant requirements;
• setup instructions and estimated times;
• in-process and final inspection requirements;
• quality characteristics, tolerances, and acceptance criteria;
• safety, environmental, and traceability requirements.

A useful process plan must be technically feasible and economically defensible. Feasibility means that the selected machine can provide adequate work envelope, power, speed range, rigidity, accuracy, and control capability. Economic defensibility means that the plan achieves the required specification at an acceptable total cost and risk.

1.1 Design information and manufacturability

The planner begins with the engineering drawing, three-dimensional model, material specification, tolerance information, surface-finish requirements, and production quantity. Features should be classified as external turning, internal turning, milling, drilling, threading, grinding, inspection, or special processing.

Design for manufacturability principles include:

• use standard sizes and readily available materials;
• avoid unnecessarily tight tolerances;
• provide adequate tool access and clearance;
• use common datums and minimize reorientation;
• avoid deep, narrow cavities where possible;
• specify surface finish only where functional performance requires it;
• design for stable workholding and economical inspection.

Tolerance is not simply a drafting detail. A tighter tolerance may require slower cutting, more stable machines, temperature control, additional measurement, and finishing operations. The planner should therefore distinguish functional requirements from historical or arbitrary specifications.

1.2 Datum structure and operation sequence

A datum is a reference from which dimensions or locations are established. A robust sequence usually follows these principles:

1. Verify material and establish a reliable initial reference.
2. Create or expose primary datums.
3. Perform operations that remove substantial material while the part is rigid.
4. Machine features that depend on established datums.
5. Perform heat treatment or stress-relief processes when required.
6. Semi-finish and finish critical features after distortion-producing processes.
7. Complete deburring, cleaning, inspection, and preservation.

The sequence must also consider precedence constraints. For example, a hole pattern may need a finished face as its locating datum, and a final grinding operation should follow heat treatment. However, rigid precedence rules should not prevent alternatives where another route provides lower cost or better flow.

1.3 Group technology and routing alternatives

Group technology classifies parts into families with similar geometry or manufacturing requirements. A family may share fixtures, tools, programs, inspection equipment, and process knowledge. This reduces planning effort and setup variation. Cellular manufacturing extends this idea by arranging machines to support a family’s flow.

A part may have several feasible routings. Route selection should compare:

2. Selecting Machines, Tools, and Cutting Parameters

2.1 Machine selection

Machine selection should be based on the most demanding feature and the complete process, not merely on the largest dimension. Important criteria include:

• maximum workpiece dimensions and mass;
• spindle speed range and available power;
• axis travel, accuracy, repeatability, and thermal stability;
• rigidity and vibration behavior;
• tool capacity and automatic tool-change performance;
• control functions, probing, data collection, and networking;
• setup accessibility, safety systems, and maintenance condition.

For a milling operation, the required spindle speed should lie within the machine range, and the machine must supply the required torque and power. For turning, chuck capacity, workholding force, tailstock or steady-rest support, and bar-feeding capability may dominate the selection.

2.2 Cutting parameters

For a rotating tool or workpiece, cutting speed is related to rotational speed by:

where v is cutting speed in m/min, D is diameter in mm, and N is rotational speed in rev/min. The equation assumes the effective cutting diameter is known and neglects small effects such as tool nose geometry.

Thus,

For milling with z teeth and feed per tooth f_z in mm/tooth, table feed is:

where V_f is feed rate in mm/min. For turning or drilling, a feed specified in mm/rev gives:

where f is feed per revolution in mm/rev.

The basic material-removal rate for a milling operation can be approximated by:

where Q is material-removal rate in mm³/min, a_p is axial depth of cut in mm, and a_e is radial width of cut in mm. This approximation is applicable when the engaged cutting width and depth are reasonably represented by constant values.

Cutting conditions are constrained by tool life, machine power, rigidity, workpiece stability, surface finish, chip control, and safety. A high cutting speed may reduce cycle time but shorten tool life or increase thermal damage. A high feed may improve productivity but increase cutting force and roughness. Parameter selection is therefore an engineering compromise.

2.3 Machining time

For a simple straight turning pass, machining time is approximately:

where t_m is cutting time in minutes, L is effective cutting length in mm, f is feed in mm/rev, and N is spindle speed in rev/min. Approach and overtravel should be included in L when they materially affect the cycle.

For a milling pass:

where L_t is total tool-center travel in mm, including approach and exit distances, and V_f is feed rate in mm/min.

The total cycle time is not just cutting time:

where t_c is cycle time, t_m is machining time, t_s is setup allocation per part, t_h is handling and loading time, t_i is inspection time, and t_d is tool-change or tool-adjustment time allocated per part. The formula assumes these elements are additive and that overlapping activities have not been double-counted.

For a batch of n parts, if setup time is incurred once per batch:

and the average time per part is:

This demonstrates why batch size affects unit cost: setup time is spread across more parts, although excessive batch size may increase inventory and lead time.

3. Quality, Capability, and Throughput

3.1 Quality at the source

Quality should be designed into the process rather than detected only at final inspection. Controls include standard work, tool presetting, first-piece approval, in-process probing, statistical process control, mistake-proofing, preventive maintenance, and clear reaction plans.

A process may be statistically stable yet incapable of meeting specification. Conversely, an apparently capable process may be unstable because of tool wear, temperature drift, or operator variation. Control charts and capability analysis should therefore be interpreted together with engineering knowledge.

For a normally distributed characteristic with process mean μ and standard deviation σ, the potential capability index is:

where USL and LSL are the upper and lower specification limits. The centered capability index is:

These indices assume a meaningful estimate of within-process variation and are most appropriate when the distribution is approximately normal and the process is stable. They do not prove that the process is stable, nor do they replace engineering judgment. A common internal target may be Cpk ≥ 1.33, but the appropriate target depends on risk, customer requirements, and industry standards.

3.2 Throughput and bottlenecks

Throughput is the rate at which accepted products leave a system. A bottleneck is a resource whose capacity limits system output. Improving a non-bottleneck may increase local utilization without increasing total production.

For a resource with available production time T_a and required processing time per part t_r, theoretical capacity is:

Units must be consistent—for example, both quantities in minutes. If demand exceeds bottleneck capacity, the process cannot meet the demand without additional time, improved methods, subcontracting, or a changed product mix.

Little’s Law relates work-in-process, throughput, and flow time in a stable system:

where WIP is average work-in-process, TH is average throughput rate, and CT is average flow time. The relationship applies to a reasonably stable system over a defined boundary and is a descriptive identity, not a scheduling method by itself. Increasing WIP can increase flow time without increasing throughput when the bottleneck is already saturated.

4. Capacity, Utilization, and OEE

Capacity analysis distinguishes calendar time from productive time. Let:

• T_c: calendar or scheduled time;
• T_d: planned downtime, such as breaks or scheduled maintenance;
• T_o: operating time after unplanned stops;
• T_i: ideal cycle time per unit;
• N: total units produced;
• N_g: good units produced.

Availability is commonly defined as:

Performance is:

Quality is:

Overall equipment effectiveness is:

OEE is a focused indicator of effectiveness relative to planned production time and ideal cycle performance. It should not be interpreted as the same thing as total factory utilization. A high OEE on a machine that is not a bottleneck may have little strategic value, while a bottleneck with moderate OEE may deserve immediate attention.

Worked Example 1: OEE

A machine is scheduled for 480 minutes. Planned breaks total 30 minutes. Unplanned downtime is 50 minutes. The ideal cycle is 0.50 min per part. During operating time, 760 parts are produced, of which 745 are good.

Planned production time is 480 − 30 = 450 min. Operating time is 450 − 50 = 400 min.

Therefore:

The OEE is approximately 82.7%. The loss structure shows that availability is the largest of the three component losses in this example; maintenance and changeover analysis should therefore be examined before simply increasing cutting speed.

5. Manufacturing Cost and Economic Decisions

5.1 Cost structure

Manufacturing cost should be separated into costs that change with volume and costs that do not change over the relevant decision range. Variable costs may include direct material, direct labor, consumable tooling, energy, and outsourced processing. Fixed or committed costs may include depreciation, salaried supervision, facility costs, and software licenses.

A useful unit-cost model is:

where C_u is unit cost, C_m is material cost per part, C_l is direct labor cost per part, C_t is tooling and consumables per part, C_s is setup cost per batch, n is batch size, C_q is expected quality-related cost per part, and C_o represents other relevant variable costs. The model is appropriate for short-run comparisons when fixed costs are either irrelevant or treated separately.

Expected quality cost can be estimated as:

where p_r, p_s, and p_w are probabilities of rework, scrap, and warranty or customer failure, while C_r, C_s', and C_w are their respective costs. Probabilities must be based on credible historical or experimental evidence.

For comparing two alternatives with different fixed and variable costs, the break-even quantity is:

where F is fixed cost, v is variable cost per unit, and alternative 1 has the lower variable cost. This equation assumes constant unit variable costs and equal product scope. It should not be used without considering capacity, learning, quality, cash flow, and risk.

Worked Example 2: Make-or-buy comparison

A component can be produced internally with a fixed annual cost of $18,000 and a variable cost of $14 per unit. A supplier offers the component at $20 per unit, with no additional internal fixed cost. Ignoring strategic and quality differences, the internal-versus-buy break-even quantity is:

Below 3,000 units, buying is economically attractive under the stated assumptions. Above 3,000 units, internal production has the lower directly attributable cost. The decision must still examine supplier reliability, intellectual property, capacity opportunity cost, incoming inspection, transport, inventory, and the reversibility of the decision.

5.2 Contribution and capacity economics

Contribution margin per unit is:

When a bottleneck limits output, contribution per bottleneck minute is more informative than contribution per unit:

where t_b is bottleneck time per unit. This measure helps prioritize products when demand exceeds bottleneck capacity, provided market demand, contractual obligations, quality, and strategic considerations are also respected.

6. Tooling, Automation, and Make-or-Buy Decisions

6.1 Tooling decisions

Tool selection should consider tool life, cutting performance, changeover time, availability, repeatability, and total cost rather than purchase price alone. A specialized tool may reduce cycle time and improve consistency but create dependence on a particular supplier or product family.

The economic tool-life problem is often expressed through a tool-life relationship such as:

where V is cutting speed, T is tool life, and n and C are experimentally determined constants for a particular tool-workpiece combination and cutting regime. The relationship is empirical, not universal. It should not be extrapolated beyond its tested range, and it does not include all costs associated with tool changes, quality loss, or machine downtime.

6.2 Automation appraisal

Automation is justified when it improves the complete system, not merely one operation. Relevant benefits may include reduced direct labor, shorter cycle time, improved repeatability, fewer handling errors, safer work, and higher available capacity. Costs include equipment, integration, programming, maintenance, training, guarding, downtime during installation, and loss of flexibility.

Automation is especially attractive when demand is stable, the product mix is limited, the process is repeatable, and labor exposure is hazardous or ergonomically unsuitable. Flexible automation is more appropriate when product variety is high but common fixtures, programs, or robot interfaces can be standardized.

6.3 Make-or-buy framework

A defensible make-or-buy decision includes:

1. Define the decision boundary and relevant time horizon.
2. Separate avoidable from unavoidable costs.
3. Compare supplier price with internal variable and avoidable fixed costs.
4. Check capacity opportunity cost and bottleneck effects.
5. Evaluate quality, delivery, confidentiality, resilience, and compliance.
6. Examine supplier financial health and continuity risk.
7. Consider learning, strategic capability, and reversibility.
8. Document assumptions and perform sensitivity analysis.

A low quoted price may be misleading if it excludes inspection, expediting, minimum order quantities, inventory, transport, or failure costs.

7. Integrated Applied Case: Illustrative CNC Housing Route

Background

The following is an illustrative case. A manufacturer produces an aluminum pump housing in annual demand of 4,000 units. The part requires a faced datum, two milled pockets, four drilled and tapped holes, and a final dimensional inspection. Two routes are being considered:

• Route A: one modern machining center with probing and automatic tool change;
• Route B: separate conventional milling and drilling operations.

Problem

Route A requires a $60,000 fixture and programming investment, but has an estimated running cost of $31 per part. Route B requires no new fixture investment and has a running cost of $43 per part. Annual demand is uncertain between 2,000 and 6,000 units. Management must decide whether Route A is justified and whether the available machining-center capacity is sufficient.

Analysis

The annual fixed investment for Route A is treated as relevant to the comparison. The break-even quantity is:

At 4,000 units, the direct annual costs are:

Route A: $60,000 + 4,000 × $31 = $184,000.

Route B: 4,000 × $43 = $172,000.

Thus Route B is cheaper at the stated volume if the $60,000 is fully attributable to the decision. At 6,000 units:

Route A: $60,000 + 6,000 × $31 = $246,000.

Route B: 6,000 × $43 = $258,000.

Route A becomes cheaper above 5,000 units. However, the result may change if Route A reduces scrap, avoids outsourcing, improves delivery, or releases labor for a more profitable bottleneck operation.

Suppose the machining center has 1,800 available production hours annually, and Route A requires 18 minutes per part including allocated handling. Its annual capacity is:

Therefore, Route A can meet 6,000 units only if the stated availability is realistic and no additional mix, maintenance, or quality losses reduce capacity. At 6,000 units it would operate at its calculated limit, leaving little protection against disruption.

Approach and solution

The recommended decision is conditional rather than automatic. The manufacturer should validate the time study with pilot production, measure first-pass yield, confirm fixture life, and test programming and tool-change assumptions. If expected demand is 4,000 units and no strategic benefits exist, Route B is economically preferable. If demand is likely to exceed 5,000 units, or if Route A provides substantial quality and capacity benefits elsewhere, Route A merits investment.

A staged approach could use Route B during market development while preparing standardized fixtures and a business case for Route A. This reduces the risk of committing capital before demand and process capability are demonstrated.

Learning points

• Break-even calculations are decision aids, not complete decisions.
• Capacity at the theoretical maximum is not the same as robust capacity.
• Quality, opportunity cost, and resilience can reverse a simple price comparison.
• Pilot data should replace unverified assumptions before capital approval.

8. Data, Digital Manufacturing, and AI

Modern machine tools can record spindle load, alarms, cycle time, tool life, probing results, energy consumption, and maintenance events. These data support condition monitoring, predictive maintenance, schedule adherence analysis, and identification of chronic losses.

Analytics should begin with a clearly defined decision. A dashboard that reports hundreds of indicators without action rules may increase attention but not performance. Data must be time-synchronized, traceable to the correct part and operation, and protected from unauthorized alteration.

AI and machine learning can help detect abnormal tool wear, classify defects from images, estimate remaining useful life, or recommend process parameters. Nevertheless, models can fail when product mix changes, sensors drift, labels are biased, or rare failure modes are absent from training data. Human approval remains essential for safety-critical parameter changes. Organizations should document data provenance, model limitations, access rights, cybersecurity controls, and accountability for decisions.

9. Challenges and Limitations

Process planning and production economics involve uncertainty. Common sources include:

• variation in material and tool behavior;
• optimistic standard times;
• unstable demand and product mix;
• hidden setup and inspection effort;
• machine downtime and maintenance delays;
• incomplete cost allocation;
• correlated quality failures;
• supplier disruption and changing energy prices.

Sensitivity analysis is therefore essential. For a proposed investment, vary demand, cycle time, scrap rate, tool cost, and downtime assumptions. A decision that remains favorable across plausible ranges is more robust than one that depends on a single optimistic forecast.

Ethically, production targets must never encourage bypassing guards, suppressing defect reporting, falsifying OEE, or shipping nonconforming parts. Metrics should reward reliable, safe, and conforming output rather than superficial utilization. Environmental analysis should include material waste, coolant, energy, tool disposal, and end-of-life considerations where relevant.

Best-Practice Decision Framework

A practical sequence is:

1. Clarify customer requirements and critical-to-quality characteristics.
2. Generate at least one technically feasible process route.
3. Identify bottlenecks, precedence constraints, and major risks.
4. Estimate time using observed data or validated standards.
5. Select machines and parameters within capability and safety limits.
6. Calculate cost using relevant, transparent assumptions.
7. Verify quality capability and inspection strategy.
8. Test alternatives through pilot runs or simulation.
9. Perform sensitivity, capacity, and risk analysis.
10. Implement, measure, review, and improve the process plan.

Chapter Summary

Process planning converts product requirements into a controlled sequence of operations, machines, tools, parameters, workholding, inspection, and documentation. Effective sequencing establishes reliable datums, preserves part rigidity, respects process precedence, and minimizes unnecessary setups and movement. Cutting speed, feed, material-removal rate, and machining time provide the foundation for estimating capacity and cost, but total cycle time must also include setup, handling, inspection, and tool-change effects.

Production analytics connects local machine behavior to system performance. Capacity is constrained by bottlenecks; Little’s Law links work-in-process, throughput, and flow time; and OEE separates availability, performance, and quality losses. Capability indices support analysis of variation but require a stable process and suitable distributional assumptions.

Economic decisions should distinguish fixed, variable, relevant, and opportunity costs. Break-even, contribution per bottleneck minute, expected quality cost, and tool-life analysis support decisions about routes, tooling, automation, and sourcing. These calculations must be supplemented by quality, resilience, safety, environmental, strategic, and ethical analysis. Digital systems and AI can improve visibility and prediction, but only when data quality, cybersecurity, model limitations, and human accountability are controlled.

Key Terms

Process plan: Documented method for converting raw material into an accepted product.

Datum: Reference feature used to establish dimensions or locations.

Routing: Ordered path of operations through a manufacturing system.

Setup time: Time required to prepare a machine, fixture, tools, and program for production.

Cycle time: Total time associated with producing one part or cycle.

Cutting speed: Relative speed between the cutting edge and workpiece at the cutting zone.

Feed rate: Relative advance of the tool or workpiece during cutting.

Material-removal rate: Volume of material removed per unit time.

Bottleneck: Resource that limits the output of a system.

Capacity: Maximum attainable output under specified assumptions and time.

Throughput: Rate of accepted output from a process or system.

Work-in-process: Items currently between start and completion of production.

Availability: Proportion of planned production time during which equipment operates.

Performance: Actual production rate relative to ideal production rate.

Overall equipment effectiveness: Product of availability, performance, and quality.

Process capability: Ability of a stable process to meet specification limits.

Contribution margin: Selling price minus variable cost.

Break-even quantity: Volume at which two alternatives have equal relevant cost.

Make-or-buy analysis: Evaluation of internal production versus external procurement.

Predictive maintenance: Maintenance planned from condition or failure prediction data.

Review Questions

Short Answer Questions

1. What information should be included in a process plan?
2. Why are datum selection and operation sequence closely related?
3. Distinguish machining time from total cycle time.
4. What are the three components of OEE?
5. Why can a high-utilization non-bottleneck fail to improve total throughput?

Long Answer Questions

1. Explain a systematic method for selecting machines, tooling, and cutting parameters for a new component.
2. Discuss how tolerance, surface finish, and inspection requirements influence process economics.
3. Explain the assumptions and limitations of capability indices C_p and C_pk.
4. Compare the economic and strategic factors involved in automation decisions.
5. Describe how a manufacturer should conduct a make-or-buy analysis under uncertain demand.

Analytical / Application Questions

1. A turning operation uses a diameter of 50 mm, a cutting speed of 100 m/min, and a feed of 0.20 mm/rev. Calculate spindle speed and machining time for an effective length of 120 mm. State the unit assumptions.
2. A machine has 420 minutes of planned production time, 35 minutes of unplanned downtime, an ideal cycle of 0.60 min, and produces 600 units of which 582 are good. Calculate availability, performance, quality, and OEE.
3. A supplier charges $26 per component. Internal production requires $45,000 fixed annual cost and $17 per component. Determine the break-even annual quantity and explain two nonfinancial factors that could change the decision.

Multiple Choice Questions with Answers

Question 1

Which sequence most generally supports reliable machining of a precision component?

A. Finish all surfaces first, then establish datums

B. Establish reliable datums before machining dependent features

C. Inspect only after shipment

D. Perform heat treatment after final inspection

Correct Answer: B. Establish reliable datums before machining dependent features

Question 2

A 50 mm diameter workpiece is cut at 100 m/min. What spindle speed is required, approximately?

A. 318 rev/min

B. 637 rev/min

C. 1,000 rev/min

D. 6,370 rev/min

Correct Answer: B. 637 rev/min

Question 3

In the equation OEE = A × P × Q, what does Q represent?

A. Quantity scheduled

B. Tool qualification

C. Quality or proportion of good units

D. Queue time

Correct Answer: C. Quality or proportion of good units

Question 4

Which resource is the bottleneck?

A. The resource with the highest purchase price

B. The resource that limits system output

C. The resource with the lowest labor rate

D. The resource with the largest floor area

Correct Answer: B. The resource that limits system output

Question 5

For a stable system, Little’s Law is expressed as:

A. WIP = TH × CT

B. WIP = TH ÷ CT

C. WIP = CT ÷ TH

D. WIP = TH + CT

Correct Answer: A. WIP = TH × CT

Question 6

If specification limits are 10 and 20 and process standard deviation is 1, what is C_p?

A. 0.60

B. 1.00

C. 1.67

D. 3.33

Correct Answer: C. 1.67

Question 7

A machine has 450 minutes of planned production time and 50 minutes of unplanned downtime. What is its availability?

A. 8.9%

B. 50.0%

C. 88.9%

D. 90.0%

Correct Answer: C. 88.9%

Question 8

When a bottleneck limits production, which measure is usually most useful for product prioritization?

A. Selling price per unit only

B. Contribution margin per bottleneck minute

C. Machine purchase price per unit

D. Total number of operations only

Correct Answer: B. Contribution margin per bottleneck minute

Question 9

An internal process has fixed cost $12,000 and variable cost $18 per unit. Buying costs $22 per unit. What is the break-even quantity?

A. 480 units

B. 2,000 units

C. 3,000 units

D. 30,000 units

Correct Answer: C. 3,000 units

Question 10

Why should AI recommendations for safety-critical machining changes receive human review?

A. AI systems never use production data

B. Models may fail under changed conditions or biased data

C. Human review always reduces capacity to zero

D. AI cannot process numerical information

Correct Answer: B. Models may fail under changed conditions or biased data