Industrial Machinery Mechanics
49-9041.00Median wage $64,520/yr439,640 employed (US)Rank #353 of 923 scored · top 38% by substitution
Repair, install, adjust, or maintain industrial production and processing machinery or refinery and pipeline distribution systems. May also install, dismantle, or move machinery and heavy equipment according to plans.
Sub-scores
0–100 · band = confidence interval from rater disagreement
Substitution32
Exposure27
Augmentation53
Substitution — the headline: capability discounted by cost, barriers and adoption.
Exposure — technical capability alone, regardless of whether anyone deploys it.
Augmentation — how much AI assists without replacing. High here + moderate substitution = a changing job, not a disappearing one.
Tasks on the substitution scale
16 rated tasks, binned by substitution score.
Position among all scored occupations
Distribution of 923 occupation scores; the marker is this occupation.
Tasks with substitution ≥ 70
6%
Run 1.0.0-draft.1 · computed 2026-08-05 · rater panel: claude-sonnet-5, claude-haiku-4-5-20251001 · intervals span rater disagreement.
Why this score
The five weighted dimensions of the composite, averaged across this occupation's tasks (importance-weighted, panel mean). Exact weights and formulas: /api/v1/methodology.
Task automatabilityw 35%28
panel mean rating 2.1/5 → substitution pressure 28/100
Technical feasibility todayw 20%24
panel mean rating 1.9/5 → substitution pressure 24/100
Cost vs. human wagew 15%29
panel mean rating 2.2/5 → substitution pressure 29/100
Adoption barriersw 20%inverted — strong barriers lower the score50
panel mean rating 3.0/5 (barrier strength) → substitution pressure 50/100
Sector adoption velocityw 10%26
panel mean rating 2.0/5 → substitution pressure 26/100
Task breakdown (16 tasks)
Substitution pressure per task, weighted by O*NET importance in the composite. Expand a task for the full rater audit trail — every rating, every model, every rationale.
Record parts or materials used and order or requisition new parts or materials, as necessary.
74CI 72–75 · exposure 75 · augmentation 75 · importance 4.0/5 · click for rater detail
Record parts or materials used and order or requisition new parts or materials, as necessary.
74| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 4/5 | Manufacturing and industrial sectors have been early adopters of inventory management automation for decades; modern cloud-based systems with AI-assisted ordering are standard practice in mid-to-large facilities and increasingly in smaller operations. |
| Sector adoption velocity | claude-sonnet-5 | 3/5 | Manufacturing and industrial maintenance sectors adopt digital inventory/CMMS tools at a moderate pace, lagging behind purely digital industries like finance or software but with growing usage of automated procurement systems. |
| Augmentation potential | claude-haiku-4-5-20251001 | 4/5 | AI assists mechanics and inventory managers by automatically flagging low stock, suggesting reorder quantities, and surfacing preferred suppliers, allowing humans to focus on exceptions and complex decisions rather than routine data entry. |
| Augmentation potential | claude-sonnet-5 | 4/5 | AI-enabled inventory systems can auto-populate usage logs, predict reorder points, and flag low stock, meaningfully reducing the mechanic's administrative burden while they retain oversight. |
| Task automatability | claude-haiku-4-5-20251001 | 4/5 | Modern inventory management systems can automatically track parts usage from sensor data or barcode scanning and trigger reorder workflows based on threshold rules, achieving substantial time savings. However, judgment about which specific parts to order (equivalents, suppliers, quantities) may still require human oversight in complex environments. |
| Task automatability | claude-sonnet-5 | 4/5 | Recording parts usage and generating reorder requisitions is a structured data-entry and inventory-management workflow that current software and AI-augmented CMMS/ERP systems handle well with minimal human input beyond verification. |
| Adoption barriers | claude-haiku-4-5-20251001 | 2/5 | Few legal or regulatory barriers prevent automation of parts tracking and ordering; most organizations already use automated systems. Integration with existing ERP systems may require some setup, but this is organizational friction rather than a hard barrier. |
| Adoption barriers | claude-sonnet-5 | 2/5 | No licensing or safety-critical sign-off is typically required for parts recording/ordering, though some organizational approval workflows for purchasing may add minor friction. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 4/5 | Integrated inventory systems spread their cost across an organization and cost significantly less per transaction than a dedicated human parts clerk. The all-in cost of AI-driven automation is well below the loaded wage of an inventory or purchasing clerk. |
| Cost vs. human wage | claude-sonnet-5 | 4/5 | Automated inventory/requisition software costs a small fraction of a mechanic's time spent on manual record-keeping and ordering, given low per-transaction software costs versus loaded technician wages. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 4/5 | Deployed ERP and inventory management systems (SAP, Oracle, NetSuite) routinely automate parts tracking and purchasing workflows in manufacturing settings. Most of the recording and requisition process is production-ready, though some edge cases (unusual parts, supplier changes) require human intervention. |
| Technical feasibility today | claude-sonnet-5 | 4/5 | Inventory management systems, CMMS platforms, and AI-driven procurement tools are widely deployed in production settings today, automating parts tracking and reorder triggers reliably at scale. |
Record repairs and maintenance performed.
69CI 65–72 · exposure 70 · augmentation 75 · importance 4.0/5 · click for rater detail
Record repairs and maintenance performed.
69| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 3/5 | Industrial maintenance is digitizing moderately fast, with CMMS (computerized maintenance management systems) and IoT adoption accelerating. However, many smaller shops and legacy operations still rely on manual logging, limiting deep, fast adoption in the sector overall. |
| Sector adoption velocity | claude-sonnet-5 | 2/5 | Industrial maintenance/manufacturing sectors are historically slower digitizers, though CMMS adoption is growing incrementally rather than rapidly with AI-native tools. |
| Augmentation potential | claude-haiku-4-5-20251001 | 4/5 | AI can greatly assist by auto-populating forms from voice or image input, suggesting part codes, and drafting maintenance summaries, allowing the mechanic to focus on accuracy and completeness rather than transcription. The human remains in control of what is recorded and can correct AI suggestions quickly. |
| Augmentation potential | claude-sonnet-5 | 4/5 | AI-assisted dictation, auto-fill templates, and summarization tools can meaningfully speed up and standardize how mechanics record their work while they remain in control of content accuracy. |
| Task automatability | claude-haiku-4-5-20251001 | 4/5 | Recording repairs and maintenance is largely a data-entry task with structured, repetitive elements (dates, part numbers, labor hours, descriptions). AI systems today can extract information from images or voice, populate forms, and generate standardized maintenance reports, delivering substantial time savings with quality equal to human entry. |
| Task automatability | claude-sonnet-5 | 4/5 | Documenting repairs is largely a structured data-entry/reporting task; voice-to-text, mobile CMMS forms, and AI transcription/summarization can handle most of this with high time savings. |
| Adoption barriers | claude-haiku-4-5-20251001 | 2/5 | No licensing or legal requirement for a human to perform data recording; the main barrier is organizational habit and preference to keep records human-validated. Customer or regulatory expectations for human sign-off on work logs may add friction, but they are not hard barriers. |
| Adoption barriers | claude-sonnet-5 | 2/5 | No licensing requirement for recordkeeping itself, though accuracy/liability concerns for maintenance records in regulated industries add mild friction. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 4/5 | AI inference for transcription, form-filling, and report generation is inexpensive per instance (pennies to cents), while a mechanic's time recording this information costs significantly more. Integration and oversight overhead are modest, yielding a strong cost advantage for the AI solution. |
| Cost vs. human wage | claude-sonnet-5 | 4/5 | Automated logging via mobile apps/voice-to-text is very cheap compared to a mechanic's time spent handwriting or typing reports, though some integration and oversight costs remain. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 4/5 | Mature products exist: voice-to-text systems, computer vision for part identification, and maintenance management software with AI-assisted form-filling are deployed in manufacturing and industrial settings. Error rates on structured data capture are low, though edge cases (unusual equipment or handwritten notes) may need human review. |
| Technical feasibility today | claude-sonnet-5 | 3/5 | CMMS software with voice dictation, templated logging, and AI summarization exists and is used in industry, but many shops still rely on manual paper/digital entry by technicians, so reliability varies by deployment maturity. |
Enter codes and instructions to program computer-controlled machinery.
59CI 30–87 · exposure 58 · augmentation 75 · importance 3.7/5 · click for rater detail
Enter codes and instructions to program computer-controlled machinery.
59| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 4/5 | Manufacturing and industrial sectors are increasingly adopting AI-assisted coding and automation workflows. Pilot and early production adoption is widespread in digitized factories; larger organizations deploy copilot-like tools for machinery programming. |
| Sector adoption velocity | claude-sonnet-5 | 2/5 | Industrial maintenance and manufacturing sectors are slower adopters of AI compared to office/professional services, with automation focused on production processes rather than mechanic programming tasks. |
| Augmentation potential | claude-haiku-4-5-20251001 | 5/5 | AI code assistants dramatically improve programmer productivity by auto-completing syntax, suggesting logic, and catching errors in real time. The human mechanic or engineer remains in control while AI substantially accelerates the programming task. |
| Augmentation potential | claude-sonnet-5 | 3/5 | AI can assist by suggesting code snippets, troubleshooting syntax errors, or referencing manuals, meaningfully speeding up the programming portion of the task while the technician retains control. |
| Task automatability | claude-haiku-4-5-20251001 | 5/5 | Programming computer-controlled machinery through code entry is a fully automatable task. AI systems can generate, review, and modify code; modern code LLMs and agents can understand machine specifications and output correct instructions at scale, meeting the ≥50% time-saving bar. |
| Task automatability | claude-sonnet-5 | 2/5 | Entering codes to program CNC or PLC-controlled machinery requires physical interaction with control panels, machine-specific calibration, and hands-on verification that current AI cannot fully replace end-to-end.' |
| Adoption barriers | claude-haiku-4-5-20251001 | 2/5 | Safety-critical machinery may require human sign-off and testing, and some organizations impose oversight requirements. However, no legal licensing requirement mandates human programming, and the automation itself faces minimal regulatory barriers to adoption. |
| Adoption barriers | claude-sonnet-5 | 3/5 | No licensing requirement specifically bars AI from generating code, but safety-critical machinery operation creates liability concerns and typically requires a qualified technician to verify and input final instructions. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 5/5 | AI inference for code generation costs pennies per task. End-to-end (inference + minimal human oversight + integration) remains orders of magnitude cheaper than a skilled mechanic's loaded wage for equivalent output. |
| Cost vs. human wage | claude-sonnet-5 | 2/5 | While AI code generation is cheap per instance, the need for skilled technician oversight, machine-specific customization, and error correction keeps effective costs comparable to human labor. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 4/5 | Code-generation products (Copilot, Claude, ChatGPT) are deployed in production and reliably assist with programming tasks. However, verification of machine-specific code and integration with specialized industrial systems introduces material error rates in some contexts, preventing a full 5. |
| Technical feasibility today | claude-sonnet-5 | 2/5 | Some AI-assisted code generation tools exist for G-code or PLC programming, but they are not deployed at scale to autonomously program and verify machinery in production maintenance contexts. |
Study blueprints or manufacturers' manuals to determine correct installation or operation of machinery.
47CI 39–56 · exposure 38 · augmentation 75 · importance 3.8/5 · click for rater detail
Study blueprints or manufacturers' manuals to determine correct installation or operation of machinery.
47| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 2/5 | Manufacturing and industrial maintenance remain relatively slow in adopting AI-driven technical document analysis; most plants still rely on human mechanics and printed manuals rather than integrated AI-assisted systems at scale. |
| Sector adoption velocity | claude-sonnet-5 | 2/5 | Industrial maintenance and manufacturing sectors have historically low digitization and slower AI tool adoption compared to information/professional services. |
| Augmentation potential | claude-haiku-4-5-20251001 | 4/5 | AI can effectively assist mechanics by rapidly searching manuals, extracting relevant diagrams, cross-referencing specifications, and flagging potential compatibility issues, significantly raising a human's productivity in planning and troubleshooting without replacing their judgment. |
| Augmentation potential | claude-sonnet-5 | 4/5 | AI can quickly summarize manuals, answer technical questions, and highlight relevant blueprint sections, meaningfully speeding up a mechanic's research phase even though physical verification remains human-led. |
| Task automatability | claude-haiku-4-5-20251001 | 2/5 | AI can extract and summarize information from blueprints and manuals, but installation and operation decisions require contextual judgment about site-specific constraints, equipment condition, and safety factors that current systems cannot fully automate end-to-end with 50% time savings at equal quality. |
| Task automatability | claude-sonnet-5 | 3/5 | Modern multimodal AI can parse blueprints and manuals, extract specs, and answer installation/operation questions, but verifying against physical machine state and ambiguous or poor-quality diagrams still requires human interpretation and judgment. |
| Adoption barriers | claude-haiku-4-5-20251001 | 2/5 | No licensing requirement mandates human review of blueprint interpretation, though liability concerns and organizational preference for experienced mechanics create some friction; the task is technically automatable without legal barriers to deployment. |
| Adoption barriers | claude-sonnet-5 | 2/5 | No licensing requirement specifically for reading manuals, but liability for incorrect machinery installation creates pressure for human verification and sign-off in industrial settings. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 4/5 | AI inference costs for document processing and OCR are minimal compared to the fully-loaded wage of an industrial machinery mechanic, creating a substantial cost advantage even accounting for integration and oversight overhead. |
| Cost vs. human wage | claude-sonnet-5 | 3/5 | AI document analysis is cheap per query, but integration with CAD/manual formats and the need for human verification of critical installation steps narrows the cost advantage to roughly comparable overall. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 3/5 | Document AI and vision models can reliably extract text and diagrams from technical manuals, but production systems show material limitations in interpreting complex spatial relationships, cross-referencing multiple documents, and validating operational correctness in real machinery scenarios. |
| Technical feasibility today | claude-sonnet-5 | 2/5 | Some engineering-document AI tools exist for technical document Q&A, but few deployed products reliably interpret mechanical blueprints for hands-on installation guidance in production maintenance settings. |
Assign schedules to work crews.
47CI 39–55 · exposure 38 · augmentation 75 · importance 3.1/5 · click for rater detail
Assign schedules to work crews.
47| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 3/5 | Manufacturing and maintenance sectors have adopted scheduling software, but adoption remains mixed—smaller shops and plants continue manual scheduling while larger operations use systems that still require supervisors to make final crew assignments. Adoption has plateaued rather than accelerating. |
| Sector adoption velocity | claude-sonnet-5 | 2/5 | Industrial maintenance and manufacturing sectors are slower adopters of AI scheduling tools compared to information/professional services, with adoption often limited to larger enterprises. |
| Augmentation potential | claude-haiku-4-5-20251001 | 4/5 | AI scheduling tools routinely assist supervisors by proposing crew assignments, flagging conflicts, and suggesting optimal sequences given constraints. A supervisor using such tools can manage more crews and handle disruptions faster than manual scheduling alone, making augmentation strong even if full automation remains impractical. |
| Augmentation potential | claude-sonnet-5 | 4/5 | AI scheduling tools can meaningfully assist supervisors by suggesting optimized crew assignments based on skills, availability, and workload, while humans retain final decision authority. |
| Task automatability | claude-haiku-4-5-20251001 | 2/5 | Scheduling work crews requires balancing multiple real-time constraints (crew availability, skill requirements, equipment status, project deadlines) and adapting to frequent disruptions. While AI can optimize static scheduling problems, the dynamic, context-dependent nature of industrial maintenance scheduling—where unexpected equipment failures and crew absences force constant replanning—exceeds what current systems reliably handle end-to-end without substantial human oversight. |
| Task automatability | claude-sonnet-5 | 3/5 | Scheduling/assignment optimization is well within reach of AI planning tools given structured data on crew availability, skills, and job priorities, though real-world variability (equipment breakdowns, urgent repairs) requires human adjustment. |
| Adoption barriers | claude-haiku-4-5-20251001 | 2/5 | No licensing or legal requirement mandates human sign-off on crew scheduling in industrial maintenance. However, operational risk (a poorly scheduled crew missing a critical failure) and union agreements around work assignment practices create modest organizational friction that slows wholesale automation. |
| Adoption barriers | claude-sonnet-5 | 2/5 | No licensing requirement to assign schedules, but organizational trust, union rules on crew assignment, and need for human judgment on urgent repairs create moderate friction. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 3/5 | Software scheduling tools have modest per-unit costs once deployed, but integration, customization, and ongoing human oversight for exception handling keep total cost-per-schedule comparable to a supervisor's time. The labor saved on routine scheduling is offset by the overhead of monitoring and correcting the system's failures. |
| Cost vs. human wage | claude-sonnet-5 | 3/5 | Scheduling software licenses plus integration costs are moderate; savings depend on scale, and small maintenance shops may not see order-of-magnitude cost reduction versus a supervisor doing this alongside other duties. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 2/5 | Scheduling optimization tools exist in enterprise software, but they typically require extensive manual configuration and human intervention to handle industrial machinery contexts with irregular failure patterns and skill dependencies. Deployed products work for stable, predictable scheduling; machinery maintenance rarely is stable or predictable, making production reliability limited. |
| Technical feasibility today | claude-sonnet-5 | 3/5 | Workforce scheduling software with optimization algorithms is deployed in many maintenance operations, but full autonomous crew assignment without supervisor override is uncommon in industrial mechanic settings. |
Examine parts for defects, such as breakage or excessive wear.
46CI 30–61 · exposure 38 · augmentation 63 · importance 4.1/5 · click for rater detail
Examine parts for defects, such as breakage or excessive wear.
46| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 4/5 | Manufacturing and industrial sectors are actively adopting automated visual inspection systems in production environments; this is a high-digitization, capital-intensive sector with strong economic incentives and demonstrable ROI driving faster-than-average AI adoption. |
| Sector adoption velocity | claude-sonnet-5 | 2/5 | Industrial maintenance and repair is a physically-oriented, lower-digitization sector where AI adoption for hands-on inspection remains at the pilot stage rather than widespread production deployment. |
| Augmentation potential | claude-haiku-4-5-20251001 | 4/5 | AI-assisted inspection tools that highlight suspected defects for human review significantly boost inspector productivity and catch rate, allowing mechanics to focus judgment on edge cases and complex parts while the system flags routine defects with high confidence. |
| Augmentation potential | claude-sonnet-5 | 3/5 | AI-enabled tools like predictive maintenance analytics, thermal imaging analysis, or AR-guided inspection checklists can help mechanics prioritize and document defects, offering real but partial productivity gains. |
| Task automatability | claude-haiku-4-5-20251001 | 3/5 | Visual inspection for defects can be partially automated using computer vision and deep learning models, but complex judgments about 'excessive wear' vs. acceptable degradation and subtle defects often require domain expertise and physical context. Current AI achieves ~50% time savings on routine defect detection with human oversight of edge cases. |
| Task automatability | claude-sonnet-5 | 2/5 | Visual/tactile inspection of mechanical parts for wear or breakage can be partially aided by computer vision, but most real-world inspection still requires physical handling, disassembly, and tacit judgment that current AI cannot fully replicate end-to-end. Time savings are modest without heavy sensor/robotic integration. |
| Adoption barriers | claude-haiku-4-5-20251001 | 2/5 | Few regulatory or licensing barriers exist for automating defect inspection itself, though quality assurance and liability frameworks may require human sign-off on critical safety-relevant decisions. Organizational friction around trust in automation and integration with existing quality systems creates modest friction. |
| Adoption barriers | claude-sonnet-5 | 3/5 | No formal licensing requirement for this specific inspection, but safety liability, mechanical judgment needs, and reliance on physical handling create meaningful organizational and practical barriers to substitution. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 4/5 | Automated visual inspection systems have low per-unit inference cost once deployed, making them substantially cheaper than human inspection labor per part examined, though initial integration and calibration carry upfront expenses that reduce the ratio for low-volume tasks. |
| Cost vs. human wage | claude-sonnet-5 | 2/5 | Deploying machine vision or sensor-based inspection systems requires significant capital and integration cost that often exceeds the marginal cost of a skilled mechanic performing spot checks, especially for varied, low-volume industrial equipment. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 3/5 | Deployed computer vision systems (e.g., industrial defect detection software) exist and perform reliably on well-defined, standardized parts in controlled environments, but material error rates remain significant for nuanced wear assessment and defects in variable lighting or complex geometries. Production use is common in manufacturing but narrow in scope. |
| Technical feasibility today | claude-sonnet-5 | 2/5 | Vision-based defect detection systems exist in manufacturing QA lines, but they are narrow, fixed-setup applications, not general mechanic-style inspection of diverse machinery parts in variable field conditions. |
Observe and test the operation of machinery or equipment to diagnose malfunctions, using voltmeters or other testing devices.
28CI 25–30 · exposure 25 · augmentation 50 · importance 3.9/5 · click for rater detail
Observe and test the operation of machinery or equipment to diagnose malfunctions, using voltmeters or other testing devices.
28| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 2/5 | Industrial manufacturing remains moderately digitized; while some large facilities deploy predictive monitoring, the adoption of autonomous diagnostic agents in production is limited and nascent, with most operations still relying on human technician rounds. |
| Sector adoption velocity | claude-sonnet-5 | 2/5 | Manufacturing and industrial maintenance sectors are slower AI adopters compared to information/finance, though predictive maintenance analytics is a growing pilot area. |
| Augmentation potential | claude-haiku-4-5-20251001 | 3/5 | AI can usefully assist mechanics by flagging sensor anomalies, suggesting common fault patterns based on diagnostic data, and accelerating reference lookups, moderately improving diagnostic speed and reducing search time for root causes. |
| Augmentation potential | claude-sonnet-5 | 3/5 | AI-powered predictive maintenance and diagnostic software can help mechanics prioritize checks and interpret sensor/voltmeter data, improving efficiency while the human still performs physical testing. |
| Task automatability | claude-haiku-4-5-20251001 | 2/5 | While AI vision systems can analyze images of equipment and interpret voltmeter readings, this task requires physical hands-on testing with devices in varied real-world conditions and troubleshooting judgment that current AI lacks end-to-end automation for. Setup, equipment handling, and decision-making under uncertainty remain heavily human-dependent. |
| Task automatability | claude-sonnet-5 | 2/5 | Diagnosis requires physical presence, sensory inspection, and hands-on testing with instruments that current AI cannot perform independently; AI can assist with data interpretation but not the physical observation/testing loop. |
| Adoption barriers | claude-haiku-4-5-20251001 | 4/5 | Safety-critical industrial machinery diagnosis carries high liability if a malfunction is missed; regulatory frameworks (OSHA, machinery directives) often require licensed technicians to sign off on safety-related diagnostics, and customer trust strongly favors human expertise in this domain. |
| Adoption barriers | claude-sonnet-5 | 3/5 | No licensing mandate specifically requires a human for diagnostics, but liability for misdiagnosed industrial equipment failures and the need for physical handling of tools creates practical barriers to full automation. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 2/5 | The cost of AI vision systems, sensor integration, data interpretation, and required human oversight remains comparable to or exceeds the hourly wage of a machinery mechanic, especially when accounting for integration and false-positive error correction. |
| Cost vs. human wage | claude-sonnet-5 | 2/5 | Deploying sensors, IoT infrastructure, and AI analytics platforms plus retaining human technicians for physical diagnosis is often costlier or comparable to a mechanic's wage for this specific task alone. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 2/5 | AI systems can assist with image analysis and sensor data interpretation, but no deployed product reliably performs the full diagnostic cycle—physical device manipulation, multi-modal troubleshooting, and contextual malfunction identification—in production environments at acceptable error rates. |
| Technical feasibility today | claude-sonnet-5 | 2/5 | Some condition-monitoring and predictive maintenance products exist that flag anomalies from sensor data, but they don't replace hands-on diagnostic testing with voltmeters or physical observation of equipment. |
Analyze test results, machine error messages, or information obtained from operators to diagnose equipment problems.
28CI 25–30 · exposure 25 · augmentation 63 · importance 3.8/5 · click for rater detail
Analyze test results, machine error messages, or information obtained from operators to diagnose equipment problems.
28| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 2/5 | Manufacturing remains fragmented across small/medium enterprises with older equipment; digital maturity is moderate to low in many facilities. While large OEMs develop proprietary diagnostics, adoption of AI diagnostic agents on the shop floor lags professional services and finance sectors. |
| Sector adoption velocity | claude-sonnet-5 | 2/5 | Manufacturing and industrial maintenance sectors are historically slower adopters of AI compared to information/finance sectors, though predictive maintenance is a growing pilot area. |
| Augmentation potential | claude-haiku-4-5-20251001 | 3/5 | AI can assist mechanics by organizing and highlighting relevant error logs and suggesting possible causes, reducing time spent searching through manuals or databases. However, the mechanic must still synthesize findings with physical inspection and domain judgment, providing moderate but meaningful productivity gain. |
| Augmentation potential | claude-sonnet-5 | 4/5 | AI-driven analytics can meaningfully help mechanics by flagging anomalies, correlating error codes with likely causes, and prioritizing what to inspect, improving diagnostic speed while the human still performs and confirms the diagnosis. |
| Task automatability | claude-haiku-4-5-20251001 | 2/5 | While AI can parse structured error messages and logs, diagnosing equipment problems requires integrating contextual knowledge of machine history, environmental factors, and operator behavior that varies by installation. Current systems lack the reliability to achieve 50% time savings at equal quality for end-to-end diagnosis without human verification. |
| Task automatability | claude-sonnet-5 | 2/5 | AI can help interpret error codes and correlate sensor data, but full diagnosis of physical machinery problems requires integrating tacit knowledge, physical inspection, and judgment that current systems cannot fully replace end-to-end. |
| Adoption barriers | claude-haiku-4-5-20251001 | 4/5 | Liability and safety concerns are substantial: equipment failures in manufacturing can cause injury or production loss, creating strong organizational preference for human expert sign-off. Regulatory frameworks in safety-critical sectors often require a licensed mechanic to validate diagnoses and sign off on repairs. |
| Adoption barriers | claude-sonnet-5 | 3/5 | No licensing requirement specifically for diagnosis, but liability for misdiagnosis leading to equipment damage or safety incidents creates moderate friction, and technicians typically must verify AI-flagged issues before acting. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 2/5 | Implementing AI diagnostic systems requires significant domain-specific training data, integration with legacy machinery interfaces, and ongoing oversight by mechanics. Total cost (inference, integration, validation) often exceeds the wage of a single mechanic diagnosis for most industrial settings. |
| Cost vs. human wage | claude-sonnet-5 | 2/5 | Deploying diagnostic AI requires sensor infrastructure, integration engineering, and ongoing calibration, so costs are often comparable to or higher than a skilled mechanic's diagnostic time, especially for smaller operations. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 2/5 | Some diagnostic tools exist (e.g., predictive maintenance platforms that flag anomalies), but they typically narrow scope to specific machine types and still require skilled human interpretation. No mature product reliably diagnoses across diverse industrial machinery with acceptable error rates in production environments. |
| Technical feasibility today | claude-sonnet-5 | 2/5 | Predictive maintenance and diagnostic-assist products exist in some industrial settings, but they are narrow in scope, often vendor-specific, and require human verification before action. |
Clean, lubricate, or adjust parts, equipment, or machinery.
26CI 16–35 · exposure 20 · augmentation 38 · importance 4.2/5 · click for rater detail
Clean, lubricate, or adjust parts, equipment, or machinery.
26| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 2/5 | Industrial machinery maintenance remains highly fragmented across small to medium firms with slow digitization; while large manufacturers experiment with condition monitoring, actual robotic automation of cleaning and adjustment tasks is still pilot-stage rather than mainstream production deployment. |
| Sector adoption velocity | claude-sonnet-5 | 2/5 | Industrial maintenance and manufacturing sectors adopt automation slowly for physical hands-on tasks, with fixed automation more common than flexible AI-driven robotic systems for varied mechanical upkeep. |
| Augmentation potential | claude-haiku-4-5-20251001 | 3/5 | Condition-monitoring AI and predictive maintenance systems can assist mechanics by identifying when service is needed and suggesting procedures, moderately improving their productivity without replacing hands-on work. |
| Augmentation potential | claude-sonnet-5 | 2/5 | AI can assist with predictive maintenance scheduling, sensor-based alerts on when lubrication is needed, or diagnostic support, but offers little direct assistance to the physical act of cleaning, lubricating, or adjusting parts. |
| Task automatability | claude-haiku-4-5-20251001 | 2/5 | While some mechanical inspection and lubrication scheduling could be partially automated with sensors and AI planning, the hands-on physical manipulation of parts requires dexterous robotic systems that are not yet mature at scale. Current AI cannot reliably perform the fine-motor adjustments and tactile feedback needed to meet the 50% time-saving threshold end-to-end. |
| Task automatability | claude-sonnet-5 | 2/5 | This requires physical manipulation, tactile feedback, and mobility in varied industrial environments that current AI systems and robots cannot perform end-to-end reliably; only narrow, fixed-routine sub-tasks in highly structured settings are automatable today. |
| Adoption barriers | claude-haiku-4-5-20251001 | 4/5 | Heavy regulatory requirements in industrial settings, safety certification needs, liability concerns around equipment failure, and the requirement that maintenance be performed by licensed or credentialed personnel create substantial legal and organizational barriers to full automation. |
| Adoption barriers | claude-sonnet-5 | 2/5 | No licensing requirement mandates a human for this specific maintenance task, but practical barriers exist around equipment variability, safety protocols, and lack of standardized robotic solutions for diverse machinery. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 2/5 | Specialized robotics systems for maintenance tasks are capital-intensive and require significant integration costs that typically exceed the loaded wage of a skilled maintenance mechanic, especially when accounting for setup, programming, and downtime. |
| Cost vs. human wage | claude-sonnet-5 | 1/5 | Physical robotic systems capable of this varied manual work require expensive custom engineering, sensors, and integration that far exceed the cost of a mechanic performing the task with hand tools and judgment. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 2/5 | Robotic systems for equipment maintenance exist in research and narrow industrial settings, but deployed products that reliably clean, lubricate, and adjust diverse machinery with high accuracy across real production environments remain rare. Most deployed solutions handle only specific, highly standardized machines. |
| Technical feasibility today | claude-sonnet-5 | 1/5 | No deployed general-purpose product performs cleaning, lubricating, and adjusting of diverse industrial machinery parts; robotic maintenance solutions remain research-stage or limited to very specific, pre-engineered applications like automated lubrication dispensers. |
Demonstrate equipment functions and features to machine operators.
22CI 14–30 · exposure 20 · augmentation 50 · importance 3.7/5 · click for rater detail
Demonstrate equipment functions and features to machine operators.
22| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 2/5 | Industrial manufacturing remains a laggard in AI/automation adoption for service and training tasks. Most plants still rely on human mechanics for direct training, and adoption of robotic or AI-driven demonstration systems is limited to large, digitally mature firms. Pilot projects exist but production deployment is sparse. |
| Sector adoption velocity | claude-sonnet-5 | 1/5 | Industrial machinery maintenance and training is a low-digitization, physical-labor sector with minimal AI agent deployment for hands-on operator training. |
| Augmentation potential | claude-haiku-4-5-20251001 | 3/5 | AI can meaningfully assist by generating animated diagrams, video clips, or interactive 3D models of equipment functions that the mechanic then uses during live demonstration. However, augmentation is constrained by the need for physical equipment access and real-time operator interaction, which remains primarily human-driven. |
| Augmentation potential | claude-sonnet-5 | 3/5 | AI can generate training materials, simulate equipment behavior, or provide AR-assisted guides that help mechanics explain features more effectively, though it doesn't replace the physical demonstration. |
| Task automatability | claude-haiku-4-5-20251001 | 2/5 | This task requires real-time interaction with physical equipment, context-specific troubleshooting, and adaptive explanations tailored to individual operators' technical knowledge. While AI could generate instructional content or video overlays, end-to-end demonstration—including hands-on operation, equipment adjustment, and responsive clarification—remains fundamentally dependent on human presence and physical manipulation. |
| Task automatability | claude-sonnet-5 | 2/5 | This requires hands-on physical demonstration, communication, and adaptive teaching on real machinery, which current AI cannot perform end-to-end; at best AI can supply supporting materials like manuals or videos.4plac |
| Adoption barriers | claude-haiku-4-5-20251001 | 4/5 | Safety regulations and liability concerns are substantial: demonstrating heavy industrial machinery carries physical risk, and organizations face legal exposure if automated systems cause injury or improper operation. Additionally, operators often prefer hands-on, human-led training for complex equipment, and OSHA compliance may require certified human instruction for certain machinery classes. |
| Adoption barriers | claude-sonnet-5 | 3/5 | No formal licensing mandates a human demonstrator, but physical presence, safety training, and equipment familiarity create practical friction against automation. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 2/5 | Setting up AI-driven demonstration systems (robotic arms, sensor integration, video infrastructure) is capital-intensive. The loaded cost per demonstration—including setup, maintenance, and oversight—currently exceeds the cost of a skilled mechanic delivering the demonstration directly, especially when factoring in equipment-specific customization. |
| Cost vs. human wage | claude-sonnet-5 | 1/5 | AI cannot substitute for the physical demonstration itself, so the human mechanic's wage remains the necessary cost; any AI role is supplementary, not a replacement, so cost comparison favors the human doing the core task. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 2/5 | No deployed products reliably perform live equipment demonstration at scale. Robotic systems exist for controlled environments but lack the flexibility to adapt to operator questions and equipment variations. AI can generate manuals or assist with video training, but autonomous or agent-based live demonstration in production settings is not yet in reliable production use. |
| Technical feasibility today | claude-sonnet-5 | 1/5 | No deployed product autonomously conducts physical equipment demonstrations to operators; this remains a human-led, in-person training activity. |
Reassemble equipment after completion of inspections, testing, or repairs.
18CI 5–30 · exposure 13 · augmentation 38 · importance 4.1/5 · click for rater detail
Reassemble equipment after completion of inspections, testing, or repairs.
18| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 2/5 | Adoption of reassembly automation in industrial maintenance is slow and limited to high-volume, standardized assembly lines. Most machinery maintenance shops remain labor-reliant and low-digitization, with pilot adoption rare in field repair and service contexts. |
| Sector adoption velocity | claude-sonnet-5 | 1/5 | Industrial maintenance and repair is a physically intensive, low-digitization sector with minimal AI/robotic adoption for hands-on mechanical reassembly tasks. |
| Augmentation potential | claude-haiku-4-5-20251001 | 3/5 | AI vision systems can assist with part-tracking and procedural guidance during reassembly, and digital work instructions can improve human efficiency. However, the hands-on nature of the task limits augmentation impact compared to knowledge-work tasks. |
| Augmentation potential | claude-sonnet-5 | 2/5 | AI can assist with documentation, checklists, or diagnostic guidance during reassembly, but offers little direct enhancement to the physical manipulation and fitting of parts. |
| Task automatability | claude-haiku-4-5-20251001 | 2/5 | Reassembly requires physical manipulation in 3D space, precise alignment, and real-time problem-solving for variances—capabilities where current AI and robotics struggle. While some standardized components in structured environments could be robotically reassembled, most industrial machinery reassembly demands dexterous coordination and adaptive judgment that current systems cannot reliably deliver end-to-end. |
| Task automatability | claude-sonnet-5 | 1/5 | Reassembling industrial machinery requires physical dexterity, fine motor manipulation of components, torque application, and adaptive handling of misaligned or worn parts—capabilities far beyond current robotics or AI systems in unstructured maintenance settings. |
| Adoption barriers | claude-haiku-4-5-20251001 | 4/5 | Safety certification, equipment warranties, and liability for incorrect reassembly create strong adoption friction. Many industrial settings require a licensed or certified mechanic to sign off on completion, and customers often demand human accountability for high-value machinery repairs. |
| Adoption barriers | claude-sonnet-5 | 3/5 | While no formal licensing mandates a human specifically for reassembly, safety-critical equipment often requires certified technicians and sign-off, and error costs (equipment failure, safety incidents) create strong organizational caution against automation. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 2/5 | Specialized robotics for assembly are capital-intensive and require extensive setup, making per-task cost high relative to a skilled mechanic's loaded wage. Integration and oversight costs further erode the economic advantage, especially for variable or low-volume reassembly work. |
| Cost vs. human wage | claude-sonnet-5 | 1/5 | There is no viable AI/robotic substitute performing this physical task, so any theoretical automation would require expensive custom robotics far exceeding a mechanic's wage. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 2/5 | No deployed product reliably performs full equipment reassembly in production settings. Specialized robotic arms exist for narrow assembly tasks in controlled factories, but general-purpose reassembly of diverse machinery post-inspection remains research-stage or pilot-only, with high failure rates on complex or variable equipment. |
| Technical feasibility today | claude-sonnet-5 | 1/5 | No deployed product autonomously reassembles industrial equipment after repair; this remains squarely a hands-on mechanical task performed by human technicians. |
Repair or replace broken or malfunctioning components of machinery or equipment.
14CI 7–21 · exposure 8 · augmentation 50 · importance 4.2/5 · click for rater detail
Repair or replace broken or malfunctioning components of machinery or equipment.
14| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 2/5 | Adoption remains slow outside high-volume manufacturing settings. Most industrial facilities rely on human mechanics for ad-hoc repairs, and the capital requirements and technical immaturity of automated repair systems limit real-world deployment. |
| Sector adoption velocity | claude-sonnet-5 | 2/5 | Industrial maintenance sectors are adopting AI mainly for predictive analytics and diagnostics, not physical repair execution, so adoption in the hands-on repair task itself remains slow and shallow. |
| Augmentation potential | claude-haiku-4-5-20251001 | 3/5 | AI can meaningfully assist mechanics through diagnostic tools, parts databases, and troubleshooting guides that accelerate decision-making and reduce downtime. However, the physical execution step limits how transformative the augmentation can be relative to more software-centric tasks. |
| Augmentation potential | claude-sonnet-5 | 3/5 | AI-based diagnostic tools, predictive maintenance systems, and AR-guided repair instructions can meaningfully assist mechanics in identifying faults and guiding repair steps, improving efficiency without replacing the physical work. |
| Task automatability | claude-haiku-4-5-20251001 | 2/5 | While AI can assist with diagnosis and parts identification, physically repairing or replacing broken machinery components requires dexterous manipulation and real-time sensory feedback that current robotics cannot reliably provide at the skill level of a trained mechanic. The task involves unpredictable physical scenarios where human judgment and adaptation remain essential. |
| Task automatability | claude-sonnet-5 | 1/5 | Diagnosing and physically repairing or replacing mechanical components requires manual dexterity, physical tool use, and situational judgment that current AI cannot perform end-to-end; robotics for general industrial repair remains far from off-the-shelf capability. |
| Adoption barriers | claude-haiku-4-5-20251001 | 4/5 | Machinery repair often requires technician certification, union involvement in certain sectors, and direct liability for failed repairs. Equipment damage from incorrect automation creates asymmetric error costs, and many industrial facilities have regulatory or contractual requirements for human sign-off on critical repairs. |
| Adoption barriers | claude-sonnet-5 | 3/5 | No strict licensing requirement mandates a human specifically, but safety liability, equipment damage risk, and the physical nature of repair work create substantial practical barriers to automation. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 1/5 | The capital cost of robot arms, integration, and safety systems capable of industrial machinery repair far exceeds the wages of a skilled mechanic, and the technology lacks the flexibility to justify this investment across typical repair scenarios. |
| Cost vs. human wage | claude-sonnet-5 | 1/5 | There is no viable AI substitute performing physical repairs, so AI cost comparison is moot; any attempted robotic solution would be far more expensive than a skilled mechanic today. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 1/5 | No deployed product can autonomously diagnose, access, and replace broken machinery components end-to-end in real industrial settings. While vision systems can aid inspection and LLMs can assist troubleshooting, the physical repair work remains firmly in the domain of specialized robotics research rather than production systems. |
| Technical feasibility today | claude-sonnet-5 | 1/5 | No deployed product autonomously repairs or replaces broken industrial machinery components; this remains beyond current robotic manipulation and diagnostic integration in production settings. |
Operate newly repaired machinery or equipment to verify the adequacy of repairs.
14CI 5–23 · exposure 8 · augmentation 38 · importance 4.0/5 · click for rater detail
Operate newly repaired machinery or equipment to verify the adequacy of repairs.
14| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 1/5 | Manufacturing and maintenance sectors show low adoption of autonomous machinery operation verification. Most adoption remains in discrete, high-volume, standardized tasks rather than the judgment-intensive verification of diverse repair work. |
| Sector adoption velocity | claude-sonnet-5 | 1/5 | Industrial maintenance and repair work is a low-digitization, physical-labor sector with minimal AI/robotic adoption for hands-on equipment verification tasks. |
| Augmentation potential | claude-haiku-4-5-20251001 | 3/5 | AI could usefully augment by monitoring sensor data during test runs, flagging anomalies, or logging performance metrics, which would assist the mechanic's verification process. However, the human operator must remain responsible for the final judgment and control. |
| Augmentation potential | claude-sonnet-5 | 2/5 | AI-enabled sensors and diagnostic software can provide supplementary data during equipment testing, but the core operate-and-verify task itself receives limited direct augmentation. |
| Task automatability | claude-haiku-4-5-20251001 | 2/5 | Operating and verifying repaired machinery requires judgment about whether repairs are adequate, which depends on subtle sensor feedback, safety assessment, and contextual knowledge of the equipment's prior state. While AI could assist with some monitoring parameters, the end-to-end task of safely operating novel equipment post-repair and making accept/reject decisions remains largely human-dependent. |
| Task automatability | claude-sonnet-5 | 1/5 | This requires physically operating repaired machinery, observing performance, and using tactile/sensory judgment to verify repair quality—no current AI system can perform physical equipment operation and verification end-to-end. |
| Adoption barriers | claude-haiku-4-5-20251001 | 4/5 | Operators must be qualified and responsible for safe machinery operation; liability for equipment damage or worker injury creates legal and insurance barriers. Regulatory frameworks (OSHA, machinery directives) expect a qualified human to verify repairs and take responsibility for safe operation. |
| Adoption barriers | claude-sonnet-5 | 3/5 | While not licensed in most jurisdictions, safety protocols, liability for equipment damage/injury, and the need for hands-on physical presence create strong organizational barriers to non-human performance. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 2/5 | Setting up AI monitoring or robotic operation would require significant capital investment, safety validation, and integration overhead. The cost per machinery verification cycle would likely exceed the hourly labor cost of a skilled mechanic, especially for diverse equipment types. |
| Cost vs. human wage | claude-sonnet-5 | 1/5 | There is no viable AI substitute for physically operating machinery, so any comparison favors the human worker entirely; robotics for this purpose is not commercially deployed. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 1/5 | No deployed autonomous system reliably operates industrial machinery to verify repairs in production environments. Robotic systems exist for specific standardized tasks, but general machinery operation verification requires embodied presence, situational judgment, and liability responsibility that current AI cannot discharge. |
| Technical feasibility today | claude-sonnet-5 | 1/5 | No deployed product operates physical industrial machinery to test repairs; this remains a physical, hands-on task performed by human technicians. |
Repair or maintain the operating condition of industrial production or processing machinery or equipment.
9CI 5–13 · exposure 0 · augmentation 38 · importance 4.3/5 · click for rater detail
Repair or maintain the operating condition of industrial production or processing machinery or equipment.
9| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 1/5 | Industrial maintenance remains largely human-performed with slow digital adoption. Most facilities rely on technician expertise and preventive schedules rather than autonomous repair systems; robotics adoption in this domain is confined to research and pilot projects, not production displacement. |
| Sector adoption velocity | claude-sonnet-5 | 2/5 | Manufacturing and industrial maintenance sectors show slower AI adoption for physical tasks compared to information-based sectors, though predictive maintenance software adoption is growing for diagnostics. |
| Augmentation potential | claude-haiku-4-5-20251001 | 2/5 | AI can assist marginally through predictive maintenance analytics, diagnostic recommendations, and documentation retrieval, but these augmentations address only planning and analysis, not the core hands-on repair work where the mechanic remains essential and AI support is limited. |
| Augmentation potential | claude-sonnet-5 | 3/5 | AI-powered predictive maintenance, diagnostic tools, and troubleshooting guides can help mechanics identify issues faster and prioritize repairs, meaningfully assisting parts of the task. |
| Task automatability | claude-haiku-4-5-20251001 | 1/5 | Industrial machinery repair requires hands-on physical manipulation, diagnosis of complex mechanical failures, and contextual judgment in unpredictable environments. Current AI lacks the embodied robotics capabilities and real-time sensorimotor feedback needed to perform this end-to-end, and no deployed system achieves the 50% time-saving threshold for the full task. |
| Task automatability | claude-sonnet-5 | 1/5 | This requires physical manipulation, diagnosis via touch/sound/sight, and hands-on repair of mechanical systems that current AI cannot perform without embodiment in capable robotics, which does not exist off-the-shelf for this domain. |
| Adoption barriers | claude-haiku-4-5-20251001 | 4/5 | Significant barriers exist: equipment liability and safety requirements typically mandate that qualified, often licensed technicians perform critical repairs; customer contracts and insurance often require human sign-off; and workplace safety regulations restrict automation in hazardous industrial environments. |
| Adoption barriers | claude-sonnet-5 | 3/5 | While not licensed in the same way as medicine or law, safety regulations, lockout-tagout procedures, and liability for equipment failure create meaningful barriers to full automation of hands-on repair work. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 1/5 | The cost of robotic systems capable of machinery repair, integration, and on-site deployment far exceeds the loaded wage of a skilled industrial mechanic. Maintenance of such systems and the infrastructure needed make automation economically unfavorable today. |
| Cost vs. human wage | claude-sonnet-5 | 1/5 | There is no viable AI substitute performing the physical repair task, so cost comparison favors the human mechanic entirely. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 1/5 | While AI can assist with diagnostics and documentation, no production system reliably performs the core repair and maintenance work autonomously. The task demands physical intervention, precise mechanical adjustments, and adaptation to site-specific machinery configurations that deployed products cannot yet handle at scale. |
| Technical feasibility today | claude-sonnet-5 | 1/5 | No deployed product autonomously repairs or maintains industrial machinery; existing systems are limited to sensor-based monitoring and alerting, not physical intervention. |
Cut and weld metal to repair broken metal parts, fabricate new parts, or assemble new equipment.
9CI 5–13 · exposure 0 · augmentation 25 · importance 3.8/5 · click for rater detail
Cut and weld metal to repair broken metal parts, fabricate new parts, or assemble new equipment.
9| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 2/5 | Manufacturing has adopted robotic welding for high-volume, standardized production (e.g., auto body shops), but industrial machinery repair and custom fabrication remain heavily manual. Adoption in repair contexts is slow due to job variety, customer expectations, and capital constraints in smaller shops. |
| Sector adoption velocity | claude-sonnet-5 | 1/5 | Industrial maintenance and skilled trades are a low-digitization, physical-labor sector with minimal AI/robotic adoption for flexible repair tasks. |
| Augmentation potential | claude-haiku-4-5-20251001 | 2/5 | AI can assist with design optimization or defect prediction, but the core cutting and welding task itself—requiring real-time control, judgment, and physical execution—sees limited augmentation from current AI tools. CAD integration helps planning, but does not materially raise the mechanic's productivity on the welding floor. |
| Augmentation potential | claude-sonnet-5 | 2/5 | AI can assist with diagnostics, part sourcing, or generating repair instructions/CAD models, but offers little direct assistance during the physical cutting/welding act itself. |
| Task automatability | claude-haiku-4-5-20251001 | 1/5 | This task requires physical manipulation of heavy machinery, precise spatial reasoning, real-time sensory feedback, and adaptive problem-solving in unstructured environments. Current AI systems cannot operate welding torches, cutting equipment, or perform the dexterous assembly work needed; it remains fundamentally a physical automation challenge beyond today's robotics deployment at scale. |
| Task automatability | claude-sonnet-5 | 1/5 | Cutting and welding metal requires physical manipulation of tools and materials in variable, unstructured environments—current AI (software-based) cannot perform this physical task, and robotic welding automation is narrow, pre-programmed, and not a general-purpose replacement for this diagnostic/fabrication task. |
| Adoption barriers | claude-haiku-4-5-20251001 | 4/5 | Safety regulations, union agreements, equipment certification, and liability for structural integrity create significant adoption friction. Welded joints on critical equipment often require licensed inspection and sign-off, and customers frequently demand human expertise and accountability. |
| Adoption barriers | claude-sonnet-5 | 3/5 | No licensing mandate universally requires a human welder, but safety codes, certification requirements (e.g., AWS welding certifications) for certain jobs, and liability for structural/safety-critical welds create meaningful friction. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 1/5 | Industrial welding and metal fabrication robots are capital-intensive, require specialized programming and maintenance, and still need human oversight and correction. The all-in cost per task remains higher than paying a skilled mechanic for most repair and small-batch fabrication work. |
| Cost vs. human wage | claude-sonnet-5 | 1/5 | Robotic welding cells cost far more than a mechanic's wage for one-off repair/fabrication work, and lack the flexibility to justify capital investment for variable tasks. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 1/5 | No deployed product reliably performs end-to-end cutting, welding, and assembly of industrial metal parts in production settings. Specialized welding robots exist but require extensive setup for each job variant and cannot handle the diagnosis and judgment inherent in repair work. |
| Technical feasibility today | claude-sonnet-5 | 1/5 | No deployed AI product performs freeform diagnostic welding/fabrication repair; robotic welding exists only in fixed, high-volume industrial settings (e.g., automotive assembly), not for ad hoc mechanic repair tasks. |
Disassemble machinery or equipment to remove parts and make repairs.
7CI 5–10 · exposure 0 · augmentation 25 · importance 4.1/5 · click for rater detail
Disassemble machinery or equipment to remove parts and make repairs.
7| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 1/5 | Industrial maintenance remains heavily labor-dependent with slow AI adoption; most facilities still rely on human mechanics, and automation of disassembly has not penetrated production settings meaningfully. |
| Sector adoption velocity | claude-sonnet-5 | 1/5 | Industrial maintenance and repair is a physical, low-digitization sector with minimal AI/robotic adoption for hands-on disassembly tasks. |
| Augmentation potential | claude-haiku-4-5-20251001 | 2/5 | AI could assist by diagnosing which parts need removal or providing step-by-step guidance, but current vision and language models offer limited practical support for the hands-on disassembly and problem-solving that defines this task. |
| Augmentation potential | claude-sonnet-5 | 2/5 | AI can assist with diagnostics, repair manuals, or guided troubleshooting information, but offers little direct help with the physical disassembly process itself. |
| Task automatability | claude-haiku-4-5-20251001 | 1/5 | Disassembling machinery requires precise physical manipulation in unstructured, variable environments—removing fasteners, managing heavy components, and handling sensitive parts. Current AI systems lack the dexterous robots and spatial reasoning needed for reliable end-to-end execution across diverse machinery types. |
| Task automatability | claude-sonnet-5 | 1/5 | This requires physical manipulation, dexterity, and situational judgment in disassembling varied machinery, which current AI systems (software-based) cannot perform end-to-end. No robotic system exists that generically disassembles diverse industrial equipment autonomously. |
| Adoption barriers | claude-haiku-4-5-20251001 | 4/5 | Physical safety liability and warranty concerns create strong barriers: improper disassembly can cause injury or machine damage. Manufacturers often require certified technicians to perform warranty-critical work, and organizational practices favor human accountability for high-cost failures. |
| Adoption barriers | claude-sonnet-5 | 3/5 | No licensing requirement specifically bars automation, but physical dexterity, unpredictable equipment conditions, and safety/liability concerns create substantial practical barriers to substitution. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 1/5 | Specialized robotic systems capable of machinery disassembly are extremely expensive to purchase, integrate, and maintain, far exceeding the loaded cost of a skilled mechanic per task completed. |
| Cost vs. human wage | claude-sonnet-5 | 1/5 | There is no viable AI/robotic substitute performing this physical task at scale, so any hypothetical automation would require expensive custom robotics far exceeding human labor costs for this task. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 1/5 | No deployed product reliably performs full disassembly and repair of industrial machinery today. While robotic arms exist for narrow, repetitive tasks in controlled settings, they cannot adapt to the variability and judgment required in real industrial maintenance work. |
| Technical feasibility today | claude-sonnet-5 | 1/5 | No deployed product performs generalized disassembly of varied industrial machinery for repair; robotics remain narrow, task-specific, and research-stage for such unstructured mechanical work. |
Related occupations — Installation, Maintenance & Repair
How to read this
A high substitution score does not mean this job disappears — it means a large share of its current tasks face replacement pressure, so the mix of tasks is likely to change. High augmentation alongside substitution typically means the occupation reorganizes around the protected tasks. Wide confidence intervals mean the rater panel disagreed: treat those scores as open questions, not verdicts.
What would change this score
New model capabilities (automatability, feasibility), falling inference costs (cost ratio), regulation and licensing shifts (barriers), and measured sector adoption (velocity) all re-enter at every index release. Each release is recomputed, versioned and kept queryable — scores are claims with a date on them, not permanent labels.