Computer-aided design (CAD) refers to the use of computer systems and software to assist in the creation, modification, analysis, or optimization of a design. CAD is widely employed across engineering, architecture, product design, and manufacturing to produce precise 2D drawings and 3D models, enabling faster iteration, better visualization, and seamless integration with downstream processes such as simulation and computer-aided manufacturing (CAM). Modern CAD tools incorporate parametric modeling, assembly management, and rendering capabilities, forming a cornerstone of digital design workflows.

1 Historical development

1.1 Early drafting systems (1960s–1970s)

1.1.1 Sketchpad and interactive graphics

The foundation of modern CAD was laid in 1963 with Ivan Sutherland’s Sketchpad system, developed at MIT. Sketchpad allowed users to draw geometric shapes on a cathode‑ray tube (CRT) display using a light pen, enabling direct manipulation of graphical elements. It introduced concepts such as constraints (e.g., perpendicularity, parallelism) and the ability to duplicate and modify objects, establishing interactive computer graphics as a practical tool for design.

1.1.2 Mainframe-based CAD systems

During the late 1960s and 1970s, CAD systems evolved from research prototypes into commercial products running on expensive mainframe computers. Companies such as Lockheed (CADAM), General Motors (DAC‑1), and Computervision (CADDS) developed systems that replaced manual drafting boards for large aerospace and automotive firms. These systems were text‑driven and required dedicated terminals, but they dramatically improved drafting speed and accuracy for complex engineering drawings.

1.2 Commercialization and growth (1980s–1990s)

1.2.1 Rise of personal computer CAD

The advent of affordable personal computers (PCs) in the early 1980s democratized CAD. Autodesk’s AutoCAD, first released in 1982, became the dominant PC‑based drafting tool by offering 2D drafting capabilities at a fraction of the cost of mainframe systems. The IBM PC platform, coupled with graphics standards such as VGA, allowed small firms and individual designers to adopt CAD. Competing products like MicroStation and Draftx also emerged, fostering a rapidly growing market.

1.2.2 Solid modeling and parametric design

The 1980s saw the shift from 2D drafting to 3D solid modeling. Systems such as PTC’s Pro/ENGINEER (1987) introduced parametric, feature‑based modeling, where designs are built from a sequence of operations (e.g., extrude, revolve) and dimensions can be changed later, automatically updating the entire model. This “parametric” approach allowed engineers to explore design variations quickly. SolidWorks (1995) brought robust solid modeling to the Windows PC platform, further expanding accessibility.

1.3.1 Cloud-based and collaborative CAD

Since the 2000s, cloud computing has enabled CAD to move from standalone installations to web‑based platforms. Tools such as Onshape (2012) and Fusion 360 (2013) allow multiple users to edit the same model simultaneously, with automatic version control and instant updates. Cloud CAD eliminates hardware constraints and facilitates remote collaboration, especially important for distributed engineering teams.

1.3.2 Integration with artificial intelligence

Recent developments incorporate artificial intelligence (AI) and machine learning into CAD workflows. AI‑assisted features include automatic part classification, suggestion of design features, and generation of alternative geometries based on performance goals (see generative design). Neural networks are also used to automate sketch recognition, dimensioning, and error checking, reducing manual input and enabling more creative exploration.

2 Core principles and workflows

2.1 2D drafting and documentation

2.1.1 Layers, linetypes, and annotations

2D drafting remains essential for producing technical drawings and manufacturing documentation. CAD systems organize drawing elements using layers, which control visibility, color, and line weight. Linetypes (continuous, dashed, dotted) represent different object types (visible edges, hidden lines, centerlines). Annotations include dimensions, notes, tolerances, and symbols, all placed with precise alignment. Modern drafting tools also generate bill‑of‑materials (BOM) tables directly from model data.

2.2 3D modeling techniques

2.2.1 Wireframe, surface, and solid modeling

Three principal approaches exist for building 3D models. Wireframe models represent objects as a collection of edges and vertices, suitable for conceptual layouts but lacking surface information. Surface modeling defines the outer skin of an object using patches (NURBS, meshes), crucial for freeform shapes in automotive and consumer products. Solid modeling represents objects as a complete, closed volume, providing mass properties and enabling downstream simulations. Many modern CAD systems combine all three.

2.2.2 Parametric and direct modeling

Parametric modeling uses features (extrusions, cuts, fillets) applied sequentially. The model’s geometry is driven by dimensions and relationships expressed in a feature tree; changes propagate automatically. Direct modeling, in contrast, allows pushing/pulling faces and moving geometry without a feature history, making it more flexible for iterative modifications and working with imported models.

2.2.2.1 Feature-based design

In feature‑based parametric CAD, each step of the design process is recorded as a feature (e.g., boss, pocket, hole, pattern). Features can be edited, reordered, suppressed, or deleted. This approach supports design intent: for example, the diameter of a hole can be linked to a reference dimension, so that changing the reference updates the hole automatically.

2.2.2.2 History-based editing

The feature tree (design history) lists all operations in chronological order. Editing a feature early in the tree rebuilds subsequent features, maintaining consistency. History‑based editing is powerful for controlled variation but can become slow or brittle for very complex models. Users can also “roll back” the tree to inspect intermediate states.

2.3 Assembly modeling and constraints

2.3.1 Mating conditions and kinematic analysis

Assembly models combine multiple parts into a single product structure. Users define relationships (mates/constraints) such as coincidence, concentricity, distance, or angle to position parts relative to one another. Constraint‑based assembly simulates real‑world joints. Kinematic analysis can then evaluate motion (e.g., rotation of a gear train, sliding of a piston) and detect interferences. Advanced tools generate exploded views and assembly animations for manuals and presentations.

2.4 Visualization and rendering

2.4.1 Realistic materials and lighting

CAD visualization goes beyond wireframe and shaded views. Modern rendering engines apply physically based materials (metallic, glass, plastic) with properties like roughness, reflectivity, and transmittance. Lighting setups (ambient, point, directional, area lights) and environment maps (HDRI) produce photorealistic images for client presentations and marketing.

2.4.2 Animation and exploded views

Animations can be created by defining keyframes for camera movement, part motion, or assembly sequences. Exploded views, often automated, show how components separate along defined vectors. These animations are used for assembly instructions, maintenance guides, and demonstration videos.

3 Major application areas

3.1 Mechanical engineering

3.1.1 Machine parts and assemblies

Mechanical CAD is the largest application domain. Engineers design individual parts (gears, shafts, brackets, pumps) and combine them into complex assemblies (engines, robots, conveyor systems). Stress‑based optimization, lightweighting, and design for manufacturing (DFM) are common considerations.

3.1.2 Finite element analysis (FEA) integration

Many CAD platforms include or link to finite element analysis (FEA) solvers. After modeling, the part is meshed, boundary conditions and loads are applied, and results such as stress, displacement, and natural frequencies are computed. Parametric studies allow automatic optimization of thickness or shape to meet strength targets.

3.2 Architecture, engineering, and construction (AEC)

3.2.1 Building information modeling (BIM)

In architecture, CAD has evolved into building information modeling (BIM). BIM software (e.g., Revit, ArchiCAD) models buildings as intelligent 3D objects (walls, doors, windows) with embedded data (material, cost, thermal performance). BIM facilitates coordination among architects, structural engineers, and MEP (mechanical, electrical, plumbing) designers, reducing clashes during construction.

3.2.2 Structural and HVAC design

Structural engineers use CAD to design steel frames, concrete reinforcement, and foundations. HVAC designers model ductwork, piping, and equipment layouts. Specialized tools (e.g., Tekla Structures, AutoPIPE) handle code‑specific calculations and fabrication‑ready drawings.

3.3 Electronics and printed circuit boards (PCB)

3.3.1 Schematic capture and layout

Electronic CAD (ECAD) tools are used for schematic capture (circuit diagrams) and printed circuit board (PCB) layout. Components (resistors, ICs, connectors) are placed on a board, connected by copper traces, and routed using auto‑routers. Design rule checks (DRC) verify clearances and manufacturability. Popular packages include Altium Designer, Eagle, and KiCad.

3.4 Industrial and product design

3.4.1 Ergonomics and aesthetic modeling

Industrial designers use CAD to create consumer products with high visual appeal and ergonomic comfort. Freeform surface tools (Class‑A surfaces) allow sculpting organic shapes. Human‑manikin modules simulate reach, visibility, and comfort for user‑centered design.

3.4.2 Rapid prototyping and 3D printing

CAD models are directly exported to STL/AMF formats for 3D printing. Rapid prototyping enables quick physical verification of form, fit, and function. Design iterations can be fabricated in hours, speeding up product development cycles.

4 Software ecosystems

4.1 General-purpose CAD platforms

4.1.1 AutoCAD and clones

AutoCAD, first released in 1982, remains the most widely used 2D drafting tool. Its .dwg file format is an industry standard. Competitors (e.g., BricsCAD, ZWCAD) offer near‑identical command sets and file compatibility, often at lower cost. AutoCAD also supports 3D modeling, but its core strength is in documentation.

4.1.2 SolidWorks, Inventor, and Solid Edge

These mid‑range parametric solid modelers target mechanical design. SolidWorks (Dassault Systèmes) is popular for its ease of use and large third‑party ecosystem. Autodesk Inventor integrates with AutoCAD and Vault for data management. Solid Edge (Siemens) offers synchronous technology (hybrid parametric/direct) and strong sheet metal tools.

4.1.3 CATIA, NX, and Creo

High‑end CAD systems serve aerospace, automotive, and complex manufacturing. CATIA (Dassault) provides advanced surfacing and systems engineering integration. Siemens NX combines CAD, CAM, and CAE in a unified environment. Creo (PTC) maintains its parametric heritage with features like behavioral modeling and additive manufacturing support.

4.2 Specialized CAD tools

4.2.1 Architectural (Revit, ArchiCAD)

Revit (Autodesk) is the leading BIM platform, supporting parametric components (families) and multi‑discipline coordination. ArchiCAD (Graphisoft) offers similar BIM capabilities with a strong focus on historic building modeling and integrated energy analysis.

4.2.2 Electronic (Altium, Eagle, KiCad)

Altium Designer is a high‑end ECAD suite with unified schematic and PCB design. Autodesk Eagle is popular for hobbyist and small‑scale designs. KiCad is a free, open‑source alternative that has gained significant adoption, offering schematic capture, PCB layout, and 3D viewer.

4.2.3 Free and open-source alternatives (FreeCAD, LibreCAD, OpenSCAD)

FreeCAD is a parametric 3D modeler comparable to SolidWorks, with a modular architecture and Python scripting. LibreCAD is a 2D drafting tool focused on .dxf compatibility. OpenSCAD uses a script‑based approach (CSG modeling), popular for generating parametric parts programmatically.

4.3 Interoperability and file formats

4.3.1 Native versus neutral formats (DWG, STEP, IGES)

Each CAD vendor uses a proprietary native format (e.g., .sldprt for SolidWorks, .prt for NX). For data exchange, neutral formats such as STEP (ISO 10303), IGES, and the industry‑standard .dwg for 2D are widely used. STEP supports 3D geometry, assemblies, and product data. IGES is older but still common for surface data.

4.3.2 Model exchange and collaboration standards

Modern standards like JT (Jupiter Tessellation) and 3D PDF facilitate lightweight visualization and collaboration without full‑featured CAD. Industry consortia (e.g., CAx‑IF) define testing protocols for interoperability. Cloud platforms often provide built‑in translators; errors can arise in complex assemblies, requiring manual cleanup.

5.1 Integration with CAM and CAE

5.1.1 Computer-aided manufacturing (CAM)

CAD models provide the geometric basis for CAM, where toolpaths (milling, turning, drilling) are generated. Integrated CAD/CAM systems (e.g., SolidCAM, Mastercam, NX CAM) allow programmers to simulate machining, detect collisions, and export G‑code for CNC machines. This tight integration shortens the design‑to‑manufacture loop.

5.1.2 Computer-aided engineering (CAE)

CAE encompasses simulation tools—structural (FEA), computational fluid dynamics (CFD), and thermal analysis—that use CAD geometry as input. Modern platforms offer embedded simulation (e.g., SolidWorks Simulation) or dedicated environments (ANSYS, Abaqus). Design‑space exploration and optimization are increasingly automated.

5.2 Generative design and topology optimization

Generative design uses algorithms to produce multiple design alternatives that meet specified performance and manufacturing constraints. The user defines loads, supports, materials, and manufacturing methods; the system iterates thousands of shapes, often producing organic, lattice‑like structures. Topology optimization, a related technique, removes material from a solid block to achieve minimum mass while maintaining strength. Both are closely linked with additive manufacturing.

5.3 Virtual and augmented reality in CAD

VR and AR headsets allow designers to immerse themselves in a 1:1 scale model, inspecting proportions, ergonomics, and spatial relationships. In VR, users can grab and edit parts, while AR overlays digital models onto the physical world for assembly guidance or client walkthroughs. Software like Gravity Sketch uses VR for intuitive 3D sculpting. Real‑time ray tracing (e.g., NVIDIA Iray) further enhances visual immersion.

5.4 Cloud-based collaboration and version control

Cloud CAD platforms (Onshape, Fusion 360, Autodesk Docs) manage design data centrally. They provide branching, merging, and change‑history features analogous to software version control (Git). Teams can work on the same model simultaneously, with each operation tracked. Cloud‑based simulation (e.g., SimScale) also reduces the need for local computing power. Data security and internet‑reliability remain concerns.