Manufacturing Processes and Design for Manufacturability

Introduction and Context

Manufacturing converts engineering designs and raw materials into usable products. The selected process affects cost, dimensional accuracy, surface quality, production rate, material utilization, environmental impact, and product reliability. A technically elegant design may be expensive or impossible to manufacture if it contains inaccessible features, excessive tolerances, unsuitable materials, or unnecessary assembly operations.

Manufacturing-process selection and design for manufacturability (DFM) must therefore be considered together. DFM asks whether a part can be produced consistently, economically, and with the required performance. Design for assembly (DFA) extends this reasoning to the joining and handling of components. In computer-aided design and computer-aided manufacturing (CAD/CAM), the digital model is not merely a geometric description; it is also the starting point for process planning, tool-path generation, inspection, costing, and production control.

Learning Objectives

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

1. Classify machining and non-machining manufacturing processes.
2. Select a process using material, geometry, accuracy, surface finish, and production-volume requirements.
3. Explain the relationships among tolerance, capability, inspection, and cost.
4. Apply DFM and DFA principles to improve part and product design.
5. Use a structured process-selection method in CAD/CAM planning.
6. Evaluate manufacturing alternatives through a practical case.

1. Manufacturing Requirements and Process Families

A process should be selected only after the functional requirements of the product are understood. Important requirements include:

• material type, hardness, ductility, melting temperature, and thermal sensitivity;
• part size, thickness, shape, and feature accessibility;
• dimensional tolerance and geometric tolerances;
• surface roughness, waviness, and appearance;
• production volume and product life;
• available equipment, tooling, skills, and inspection facilities;
• material utilization, lead time, safety, and environmental constraints.

Manufacturing processes may be grouped as follows:

A part commonly requires a process chain rather than one operation. For example, a steel shaft may be forged, normalized, turned, heat-treated, ground, and inspected. Process planning must consider how an earlier operation influences later datum surfaces, residual stress, distortion, and stock allowance.

2. Machining Processes

Machining is a subtractive process in which a cutting tool, abrasive, or electrical discharge removes unwanted material. It is especially valuable when close tolerances, accurate locations, complex surfaces, or good finishes are required.

2.1 Turning

Turning rotates the workpiece while a cutting tool removes material. It is suitable for cylindrical, conical, threaded, and faced features. CNC turning centres may include live tooling, automatic tool changers, and bar feeders.

For a workpiece rotating at N revolutions per minute and having diameter D in millimetres, cutting speed is:

where v is cutting speed in metres per minute. The factor 1000 converts millimetres to metres. This relation assumes that the stated diameter represents the effective cutting diameter and that rotational speed is constant.

The theoretical feed distance per revolution is f millimetres per revolution. For turning, feed rate is:

where F is feed rate in millimetres per minute. Actual surface roughness also depends on tool nose radius, tool geometry, workpiece material, machine rigidity, vibration, and built-up edge; therefore, a simple feed equation should not be treated as a complete roughness prediction.

2.2 Milling

In milling, a rotating multi-tooth cutter removes material from a stationary or moving workpiece. Face milling produces planar surfaces, while peripheral milling, slotting, pocketing, and contour milling create other features. CNC milling is highly flexible but may require multiple setups and careful control of tool access.

For a cutter with z teeth, feed per tooth f_z, and spindle speed N, the table feed is:

The equation assumes that all teeth engage as intended and that feed per tooth is specified in millimetres per tooth. It is used for planning, not as a guarantee of tool life or surface finish.

2.3 Drilling and Hole-Making

Drilling is economical for producing holes, but the resulting accuracy and finish may not satisfy all functional requirements. Reaming improves hole size and finish; boring corrects location or enlarges an existing hole; tapping produces internal threads. A design should avoid deep, small-diameter holes when possible because they increase tool deflection, chip-evacuation difficulty, and breakage risk.

2.4 Grinding and Abrasive Processes

Grinding uses bonded or coated abrasive grains. It is generally selected for hardened materials, close dimensional tolerances, and fine surface finish. Grinding can generate heat and residual stress, so coolant, wheel selection, dressing, and material removal rate must be controlled.

2.5 Electrical and Thermal Machining

Electrical discharge machining (EDM) removes electrically conductive material through controlled spark discharges. It can produce narrow slots, sharp internal corners, and complex cavities in hardened materials, but it is slower than conventional milling and cannot process electrical insulators. Wire EDM cuts profiles through a workpiece; sinker EDM forms cavities using an electrode.

Laser, water-jet, and plasma cutting are often used for sheet and plate. Their suitability depends on thickness, heat-affected zones, kerf width, taper, edge quality, and material response.

3. Non-Machining Processes

3.1 Casting

Casting forms a part by pouring or injecting liquid material into a mould. It is advantageous for complex external shapes, internal cavities, large components, and low-to-high production volumes depending on the moulding method. Sand casting has relatively low tooling cost but lower accuracy and surface quality. Die casting provides high production rates and good repeatability for suitable nonferrous alloys, but tooling cost is substantial.

Casting designs require uniform wall thickness where possible, suitable draft angles, fillets, feed systems, and control of shrinkage and porosity. Sudden changes in section thickness can cause hot spots and defects.

3.2 Forging and Forming

Forging produces strong parts by plastic deformation, often with favourable grain flow. It is suitable for high-strength components such as connecting rods, gears, and aircraft fittings. Open-die forging supports large parts and low volumes; closed-die forging provides repeatable shapes at greater tooling cost.

Sheet-metal operations include blanking, bending, stamping, and deep drawing. Bend radii, bend allowance, springback, grain direction, and minimum flange dimensions must be considered. A design that requires many specialized forming stages may be unsuitable for low production volume.

3.3 Injection Moulding

Injection moulding is widely used for thermoplastics and some thermosets. Molten polymer is injected into a mould, cooled or cured, and ejected. It provides high production rates and repeatable complex shapes, but moulds are expensive. Uniform walls, adequate draft, ribs instead of excessive solid sections, appropriate bosses, and controlled shrinkage improve results.

3.4 Additive Manufacturing

Additive manufacturing builds parts from a digital model. It is valuable for prototypes, customized products, internal lattice structures, topology-optimized forms, and geometries difficult to machine. Limitations include build orientation, anisotropy, support structures, surface roughness, residual stress, material qualification, and post-processing. A printed part may still require machining of critical interfaces.

3.5 Joining Processes

Welding is efficient for permanent metal assemblies but may cause distortion, residual stress, and heat-affected zones. Bolted joints facilitate disassembly but require holes, access, fasteners, and torque control. Adhesive bonding distributes load over an area and joins dissimilar materials, but surface preparation, cure conditions, temperature, and long-term environmental exposure are important.

4. Selecting a Process

Process selection is a constrained engineering decision rather than a search for one universally best method. A useful sequence is:

1. Define functional features and quality requirements.
2. Identify feasible process families for the material and geometry.
3. Eliminate processes that cannot meet size, tolerance, finish, or safety requirements.
4. Estimate tooling, machine, material, labour, inspection, and post-processing costs.
5. Compare production rate, flexibility, risk, environmental impact, and available capability.
6. Validate the choice through prototypes, simulations, trial production, or supplier consultation.

A qualitative selection matrix can make assumptions visible. Scores should be defined before use; for example, 1 may mean poor suitability and 5 excellent suitability. Weighted scoring is useful when priorities are explicit. If criterion weights are w_i and process scores are s_i, with weights summing to one, the total score is:

This is a decision aid, not an objective law. The result is sensitive to weights and scoring judgments, and a high score cannot compensate for a process that violates a mandatory requirement.

The economic choice changes with volume. A simple total-cost model is:

where C_total is total cost, C_fixed is fixed cost such as tooling and programming, C_variable is variable cost per part, and n is production quantity. This model assumes constant variable cost and excludes financing, learning effects, scrap variation, and capacity constraints. The break-even quantity between processes A and B is:

provided the denominator is nonzero and the compared cost models are valid. If the result is negative, one process is cheaper in both fixed and variable components within the model.

Worked Example: Volume and Process Choice

An aluminium housing can be CNC machined with fixed cost $1,200 and variable cost $18 per part, or die-cast with fixed tooling cost $18,000 and variable cost $5 per part. Equating costs:

1200 + 18n = 18000 + 5n

Thus, 13n = 16800, giving n = 1292.3. Since production quantity is normally a whole number, die casting becomes economically preferable above approximately 1,293 parts, assuming both processes meet the required quality and capacity is available. At 500 parts, machining costs $10,200, whereas die casting costs $20,500. At 2,000 parts, machining costs $37,200, whereas die casting costs $28,000.

5. Accuracy, Surface Quality, and Process Capability

Tolerance is the permitted variation in a dimension. A tolerance should be no tighter than function requires. Tight tolerances increase machining time, tool requirements, inspection effort, rejection risk, and sometimes the need for grinding or special temperature control.

A process capability index compares the spread and centring of a process with specification limits. For a stable process with approximately normal output, the capability ratio is:

and the centred capability index is:

Here, USL and LSL are upper and lower specification limits, μ is the process mean, and σ is the process standard deviation. These indices are meaningful only when the process is statistically stable and the measurement system is adequate. A high C_p with low C_pk indicates that the process variation may be small but the mean is off-centre.

Surface roughness is commonly reported by R_a, the arithmetic mean deviation of a measured profile. It is not identical to waviness or visual appearance. Surface requirements should be specified only where they influence friction, sealing, fatigue, wear, bonding, or appearance.

Datums establish the reference framework for manufacturing and inspection. Functional datums should be selected so that setup, machining, assembly, and measurement reference the same surfaces whenever practical. Poor datum selection can create tolerance stack-up and inconsistent assembly.

6. Design for Manufacturability

DFM is the systematic design of products so that they can be manufactured with reliable quality and reasonable cost. Important principles include:

• use standard materials, stock sizes, tools, fasteners, and purchased components;
• minimize the number of unique parts and operations;
• avoid unnecessary tight tolerances and finishes;
• provide tool access and adequate clearance;
• use consistent wall thickness where casting or moulding is involved;
• design radii compatible with cutters and forming tools;
• avoid deep narrow pockets and fragile projections;
• use self-locating and self-fixturing features where possible;
• design inspection datums into the part;
• reduce setups and repositioning;
• provide chip clearance and coolant access;
• account for draft, shrinkage, springback, distortion, and tool deflection;
• ensure that the CAD model communicates material, tolerance, finish, and revision requirements unambiguously.

Machining-Specific DFM

Internal corners produced by rotating cutters normally contain a radius. A sharp internal corner may require a small tool, multiple passes, EDM, or a redesign. Pocket depth should be limited relative to width, and thin walls should be avoided because they can deflect under cutting forces. Holes should be aligned with available spindle directions when possible. A single setup is generally preferable to several setups, but a theoretically single-setup design may be inferior if it requires an expensive fixture.

Casting and Moulding DFM

Draft allows a part to leave a mould without damage. Ribs can increase stiffness without creating thick, defect-prone sections. Bosses should be supported and connected to walls with suitable transitions. Parting lines should be placed where flash and witness marks are acceptable. Designers must distinguish cosmetic surfaces from functional surfaces so that mould complexity is justified only where needed.

Sheet-Metal DFM

Designers should consider minimum bend radius, bend sequence, tooling access, hole-to-edge distance, bend relief, grain direction, and springback. Features that can be produced in the same die or setup reduce cost. The flat pattern must be generated using material-specific bend data rather than an assumed universal value.

7. Design for Assembly

DFA reduces assembly time, errors, and equipment requirements. A product should ideally have a small number of parts, a clear assembly direction, easy handling, and features that prevent incorrect orientation. Parts should be easy to grasp, feed, locate, and join.

Useful DFA questions include:

1. Does this part move relative to every other part?
2. Must it be made from a different material?
3. Must it be separate for service, replacement, or manufacture?
4. Can two parts be combined without impairing function?
5. Can the assembly be completed from one principal direction?

Snap fits, captive fasteners, symmetry, keyed interfaces, and visual error-proofing can improve assembly. However, an integrated part may be more difficult to repair or recycle, so lifecycle requirements must be considered.

8. CAD/CAM Integration and Digital Process Planning

The CAD model should support manufacturing rather than merely display shape. Product data should include material, units, coordinate systems, datums, tolerances, surface specifications, revision status, and manufacturing notes. CAM planning then determines stock, setups, fixtures, tools, cutting conditions, tool paths, verification, and inspection operations.

A robust digital workflow is:

CAD model → manufacturability review → process plan → fixture and tool design → CAM programming → simulation and collision checking → machining → inspection → feedback to design.

Design reviews should include machinists, process engineers, quality engineers, suppliers, and assembly personnel. Digital simulation can identify tool collisions, excessive axis motion, unreachable features, remaining stock, and poor surface scallop patterns. It cannot by itself guarantee material behaviour, fixture rigidity, tool life, or operator safety.

AI-assisted manufacturing can suggest process parameters, detect anomalies, classify defects, and optimize tool paths. Its output must be validated against approved materials, machine limits, cybersecurity controls, traceability requirements, and human engineering judgment. Training data may encode bias or omit rare failure modes; therefore, AI should support—not replace—qualified process approval.

9. Case Study: Redesign of an Aluminium Sensor Bracket

The following is an illustrative case based on common industrial practice.

Background

A small aluminium sensor bracket was originally machined from billet. It contained six drilled holes, two deep pockets, four separate fasteners, and three tight positional tolerances. Annual demand was 2,400 units.

Problem

The bracket required three machining setups, generated substantial material waste, and had a high rejection rate caused by thin-wall deflection and hole-location variation. Assembly technicians also installed two similar brackets in the wrong orientation.

Analysis

The design team separated functional requirements from historical preferences. Only two mounting holes required close positional tolerance. The remaining holes were clearance holes. The deep pockets contributed little stiffness but increased machining time. The four fasteners could be reduced to two by adding locating tabs and a keyed interface.

Alternative processes were considered. Investment casting could reduce waste and create the shape, but tooling and casting qualification would be significant. A revised CNC-machined design was selected for the current volume because it retained flexibility and avoided new casting tooling. A cast version was retained as a future option if demand increased.

Approach and Solution

The redesign used:

• two primary datums corresponding to functional mounting surfaces;
• larger internal radii compatible with standard cutters;
• shallow pockets and ribs to maintain stiffness;
• two dowel-style locating features and two fasteners;
• distinct asymmetrical orientation features;
• relaxed tolerances on nonfunctional holes;
• one principal machining setup plus a simple finishing setup;
• inspection of critical features using a defined datum reference frame.

CAM simulation confirmed tool access and reduced non-cutting motion. A pilot batch was measured before release.

Outcome

The number of setups fell from three to two, assembly fasteners from four to two, and the pilot rejection rate decreased substantially. Material waste and machining time were reduced. The exact savings would depend on machine rates, tooling, labour, and volume, but the central improvement came from removing unnecessary requirements rather than merely increasing cutting speed.

Learning Points

The case demonstrates that DFM is not simply a machining department activity. Functional analysis, tolerance rationalization, datum planning, assembly simplification, and CAD/CAM verification must occur together. It also shows why process choice should be revisited when volume, material, or design requirements change.

10. Challenges, Limitations, and Professional Responsibilities

No process-selection method eliminates uncertainty. Supplier capability may differ from published nominal capability. Tool wear, material batches, thermal drift, measurement uncertainty, and operator practice influence results. Cost models may omit downtime, scrap, maintenance, energy, and end-of-life effects.

Engineers must not specify a process solely because it is familiar or because its equipment is available. Safety-critical parts require documented material traceability, qualified processes, calibrated inspection, controlled revisions, and appropriate standards. Welding, pressure-containing components, medical products, and aerospace parts may require specialized codes and certification.

Environmental responsibility includes minimizing scrap, selecting recyclable materials where feasible, reducing hazardous fluids and emissions, managing waste, and considering energy over the product life cycle. Design choices should not transfer environmental harm from manufacturing to difficult disassembly or disposal.

Best-Practice Checklist

Before releasing a design, ask:

• Is every tight tolerance functionally justified?
• Can the selected process make the geometry repeatedly?
• Are datums consistent with function, setup, and inspection?
• Are tools, fixtures, and measuring equipment accessible?
• Has production volume been included in the economic comparison?
• Are material, heat treatment, finish, and revision requirements explicit?
• Can parts be assembled, serviced, and oriented correctly?
• Has the CAM program been simulated and independently checked?
• Have safety, environmental, supplier, and regulatory requirements been addressed?

Chapter Summary

Manufacturing-process selection depends on material, geometry, accuracy, surface quality, production volume, equipment, cost, risk, and environmental requirements. Machining provides flexibility and precision, while casting, forming, moulding, joining, and additive processes may offer superior productivity or material efficiency for appropriate applications. Cost comparisons must distinguish fixed and variable costs and recognize the effect of production quantity.

DFM reduces unnecessary complexity, excessive tolerances, difficult tool access, waste, and process variation. DFA reduces part count, handling difficulty, orientation errors, and joining effort. Datums, process capability, surface specifications, CAD data quality, CAM simulation, inspection, and feedback form an integrated digital manufacturing system. The best design is not simply the easiest part to machine; it is the product that satisfies its function with reliable quality, economical production, safe operation, and responsible lifecycle performance.

Key Terms

Manufacturing process: A method used to transform material into a product or component.

Machining: Subtractive manufacturing using a cutting tool or abrasive.

Casting: Shaping a part by solidifying liquid material in a mould.

Forging: Deforming material under compressive force to create a shape.

Forming: Changing material shape primarily through plastic deformation.

Additive manufacturing: Producing a part layer by layer from digital data.

Tolerance: Permitted variation in a specified dimension or feature.

Surface roughness: Fine-scale irregularity of a surface profile.

Datum: A reference point, axis, or plane used for design, manufacture, or inspection.

Process capability: The ability of a stable process to satisfy specification limits.

DFM: Design for manufacturability; designing products for efficient, repeatable production.

DFA: Design for assembly; designing products for simple and reliable assembly.

Setup: A condition in which a workpiece is located and held for an operation.

Fixture: A device that locates and supports a workpiece.

Tool access: The ability of a manufacturing tool to reach a required feature.

Fixed cost: Cost that is largely independent of the number of parts produced.

Variable cost: Cost that changes approximately with the number of parts.

Process chain: The ordered sequence of manufacturing and inspection operations.

Tolerance stack-up: Combined variation resulting from several dimensions or features.

Digital thread: Connected product and process information across the lifecycle.

Review Questions

Short Answer Questions

1. What factors should be considered when selecting a manufacturing process?
2. Distinguish machining from non-machining processes.
3. Why are unnecessary tight tolerances costly?
4. What is the purpose of a datum?
5. State two principles of design for assembly.

Long Answer Questions

1. Compare turning, milling, grinding, and EDM with respect to geometry, accuracy, material, and production use.
2. Explain how production volume influences the choice between machining and tooling-intensive processes.
3. Discuss DFM principles for cast, moulded, and machined components.
4. Explain the meaning and limitations of C_p and C_pk.
5. Describe an integrated CAD/CAM workflow from design release to inspection feedback.

Analytical / Application Questions

1. A component is required in quantities of 300, 2,000, and 20,000 per year. Propose a process-selection strategy for each volume and identify information needed before making the final decision.
2. A machined bracket contains deep narrow pockets, sharp internal corners, and ten different hole sizes. Redesign it using DFM and DFA principles, explaining each change.
3. A process has USL = 25.10 mm, LSL = 24.90 mm, μ = 25.04 mm, and σ = 0.02 mm. Calculate C_p and C_pk, then interpret whether variation or process centring is the dominant concern. State the assumptions required for your interpretation.

Multiple Choice Questions with Answers

Question 1

Which process is generally most suitable for producing a close-tolerance cylindrical shaft from billet?

A. CNC turning

B. Sand casting only

C. Manual adhesive bonding

D. Deep drawing

Correct Answer: A. CNC turning

Question 2

In the relation v = πDN/1000, what does v represent when D is in millimetres and N is in revolutions per minute?

A. Feed per tooth in millimetres

B. Cutting speed in metres per minute

C. Power in kilowatts

D. Material-removal rate in cubic millimetres per minute

Correct Answer: B. Cutting speed in metres per minute

Question 3

Which design change most directly improves machinability of an internal pocket?

A. Specify a perfectly sharp internal corner

B. Use an internal radius compatible with the cutter

C. Reduce tool access to the pocket

D. Increase the number of unique cutter sizes

Correct Answer: B. Use an internal radius compatible with the cutter

Question 4

Which process commonly uses a mould with high initial tooling cost but is economical at high volumes for thermoplastic parts?

A. Injection moulding

B. Hand filing

C. Wire brushing

D. Single-piece forging by hand

Correct Answer: A. Injection moulding

Question 5

In the total-cost model C_total = C_fixed + nC_variable, what does C_fixed represent?

A. Cost that varies with every part

B. Cost independent or largely independent of production quantity

C. Only the cost of raw material

D. Only inspection rejection cost

Correct Answer: B. Cost independent or largely independent of production quantity

Question 6

A low C_pk combined with a relatively high C_p most strongly suggests that the process is:

A. Precisely centred and incapable because of excessive spread

B. Potentially capable in spread but poorly centred

C. Unmeasurable because no specification limits exist

D. Guaranteed to be stable

Correct Answer: B. Potentially capable in spread but poorly centred

Question 7

Which is a primary DFA objective?

A. Increase the number of fasteners

B. Make component orientation less obvious

C. Reduce part count and assembly errors

D. Require a separate setup for every component

Correct Answer: C. Reduce part count and assembly errors

Question 8

Why may grinding be selected after heat treatment?

A. It is used to produce liquid metal

B. It can provide close dimensions and fine finish on hardened material

C. It eliminates the need for inspection in all cases

D. It is suitable only for soft polymers

Correct Answer: B. It can provide close dimensions and fine finish on hardened material

Question 9

Which statement about weighted process-selection scores is correct?

A. They replace all engineering judgement

B. They are valid even when a process violates a mandatory requirement

C. They depend on defined criteria, weights, and scoring assumptions

D. They always identify the lowest-cost process

Correct Answer: C. They depend on defined criteria, weights, and scoring assumptions

Question 10

Which feature most improves assembly error-proofing?

A. Two identical orientations with no locating features

B. An asymmetrical keyed interface

C. Unlabelled interchangeable fasteners

D. Hidden access to all joining operations

Correct Answer: B. An asymmetrical keyed interface