How to Evaluate and Compare Exoskeletons: A Practical Buyer’s Guide (2026)

Published: 30 July 2026

Last reviewed: 30 July 2026

Audience: Industrial buyers, healthcare organisations, rehabilitation teams, procurement teams, safety and ergonomics professionals, researchers, distributors and other organisations evaluating exoskeleton technologies.

Editorial position: Exoskeleton Index does not treat exoskeletons as interchangeable products or as guaranteed solutions. The correct question is not “Which exoskeleton is best?” but “Which device, if any, is appropriate for this task, user group, environment and regulatory context?”

Executive summary

Exoskeleton procurement should begin with the task and the user, not with a product catalogue.

The evidence base is growing, but it remains uneven. A 2024 systematic review identified 49 occupational exoskeleton studies published between 2014 and 2024 and found that much of the evidence still comes from laboratory evaluations rather than long-term workplace deployments.[1] A separate review of quality and productivity outcomes included 15 studies and found that none directly evaluated the economic implications of occupational exoskeleton use.[2] This means buyers should be cautious about guaranteed productivity, injury-prevention or return-on-investment claims.

A defensible evaluation process should therefore:

1. Define the physical problem and task before selecting a device.

2. Separate industrial, medical, rehabilitation and consumer use cases.

3. Match the device to the exact posture, motion, load and environment.

4. Evaluate fit across the real user population, not only an “average” worker.

5. Review evidence by task and outcome, not only by product marketing.

6. Assess new risks introduced by the device.

7. Test usability, comfort, compatibility and worker acceptance.

8. Review training, maintenance, support and lifecycle costs.

9. Run a controlled pilot before wider deployment.

1. Document the reasons for selection, rejection and scale-up.

—

1. Start with the problem, not the product

A buyer should be able to describe the problem in operational terms before discussing brands or models.

Examples of useful problem statements include:

• Repeated overhead fastening for 20 to 40 minutes per cycle.
• Trunk flexion during low-level assembly.
• Frequent lifting from floor level to waist height.
• Static or semi-static crouching.
• Assisted gait training under clinical supervision.
• Personal mobility support for a defined user group.
• Tool holding in constrained environments.
• Load carriage over uneven terrain.

A weak problem statement is simply “reduce injuries” or “improve productivity.” These outcomes are too broad to guide selection.

For occupational applications, exoskeletons should also be considered within a wider ergonomics and risk-control programme. NIOSH places elimination, substitution and engineering controls above administrative controls and personal protective equipment in the hierarchy of controls.[3] Exoskeletons should not be used to avoid correcting a badly designed task when the hazard can reasonably be removed or redesigned.

Questions to answer first

• What movement, posture or load creates the problem?
• How often does it occur?
• How long does each exposure last?
• Is the task predictable or highly variable?
• Can the task be redesigned without a wearable device?
• What outcome needs to improve: muscle demand, discomfort, endurance, quality, recovery, mobility or another measure?
• What would count as failure?

—

2. Decide which regulatory and use-case branch applies

“Exoskeleton” is not one regulatory or commercial category.

Occupational and industrial systems

These devices are generally evaluated as workplace technologies intended to support workers during tasks such as lifting, overhead work, tool handling, bending or crouching.

As of July 2026, ISO 25563 remains under development. The draft covers the integration of wearable physical-assistance devices in work situations, including needs definition, selection, assessment and deployment. It explicitly excludes medical, rehabilitation, games, sport and leisure applications.[4]

ASTM Committee F48 has developed a growing portfolio covering terminology, ergonomics, training, task performance, environmental conditions and risk. Relevant examples include:

• ASTM F3444/F3444M for user training.
• ASTM F3474 for functional ergonomic parameters and test metrics.
• ASTM F3518 and F3519 for quantitative measures and reporting.
• ASTM F3527 for contextual risk assessment.
• ASTM F3688 for potential ergonomic risks.
• ASTM F3749 for logistics assessment.
• ASTM F3750 for tool-handling assessment.[5][6]

These standards do not remove the need for task-specific evaluation. ASTM F3527 states that no general certification programme currently guarantees that a device has been assessed for every real working context.[6]

Medical and rehabilitation systems

In the United States, a powered lower-extremity exoskeleton used for medical purposes is a Class II prescription device under 21 CFR 890.3480 and typically follows the 510(k) route.[7]

In the European Union, the Medical Device Regulation applies according to the manufacturer’s intended medical purpose. The manufacturer’s stated intended purpose in labelling, instructions, marketing and clinical evaluation is therefore central.[8]

A hospital or rehabilitation centre should not evaluate a medical exoskeleton using only an industrial buyer checklist. Clinical governance, patient eligibility, supervision, contraindications, training, infection control, regulatory status and clinical evidence require separate review.

Consumer, sport and personal mobility products

These can sit outside medical or occupational frameworks depending on intended use and claims. Buyers should verify:

• Whether the product is marketed as a medical device.
• Whether it is intended for personal assistance, recreation or performance.
• What local product-safety rules apply.
• Whether the manufacturer has local support, warranty and liability arrangements.

—

3. Build a task profile before comparing devices

Exoskeleton effectiveness is highly task-dependent.

A systematic review of industrial exoskeleton evaluation proposed three core dimensions: the evaluation metrics used, the target task and the body posture adopted.[9] This is a useful procurement principle because the same device can perform well in one activity and poorly in another.

Minimum task-profile fields

Do not rely on the “main task” alone

Workers often move between supported and unsupported activities. A shoulder-support device may help during overhead work but interfere with reaching below the waist, entering a vehicle or carrying materials. A back-support device may reduce trunk-muscle demand during bending but become restrictive during twisting or agile movement.

An updated review of back-support exoskeletons found general reductions in back-muscle activity and spinal loading, but also reported reduced performance in tasks requiring agility, moderate user satisfaction and possible increases in abdominal or lower-limb muscle activity.[10]

The procurement question is therefore not only:

Does the device assist the target task?

It is also:

What happens during every movement the worker must perform before, during and after that task?

—

4. Match the support architecture to the task

Passive systems

Passive exoskeletons use springs, elastic elements, gas springs or mechanical structures rather than powered actuation.

Potential advantages:

• Lower weight and complexity.
• No battery charging.
• Simpler maintenance.
• Easier deployment in some environments.
• Lower acquisition cost in many cases.

Potential limitations:

• Assistance often depends strongly on posture.
• Resistance may be felt outside the intended task.
• Support may be less adaptable across task phases.
• Forces can be transferred to other body regions.

Powered or active systems

Powered systems use motors, actuators and control systems.

Potential advantages:

• Assistance can be modulated.
• Support may adapt across movement phases.
• Higher assistance may be possible.
• Some systems can collect operational data.

Potential limitations:

• Higher weight and complexity.
• Charging and battery management.
• More demanding maintenance.
• Greater control, software and failure-mode considerations.
• Higher training and service requirements.

A controlled study involving 15 men and 14 women compared three back-support devices during manual-handling tasks with a 15 kg load. The passive device reduced activity in selected muscles by 12% to 27%, while the two active devices produced reductions of 7% to 62% across the measured muscles. The results also showed that effects depended on device design and task type, and that active systems could modify trunk kinematics.[11]

These figures should not be used as universal performance claims. They demonstrate why buyers must compare devices under the intended task conditions.

Soft exosuits versus rigid structures

Soft systems may offer greater freedom and lower bulk, while rigid systems may provide clearer load paths or stronger support. Neither is automatically superior. The correct choice depends on:

• Where loads are transferred.
• Required freedom of movement.
• Interface pressure.
• PPE compatibility.
• Need for structural support.
• Donning, doffing and adjustment.
• Cleaning and maintenance.

—

5. Treat fit as a safety and performance variable

Fit is not a cosmetic issue. It can affect assistance, comfort, movement, pressure, stability and willingness to use the device.

Evaluation literature identifies fit, mobility, balance, comfort and task efficiency as core measurement areas.[12]

Buyers should test

• Height range.
• Torso, waist, hip, thigh and limb dimensions.
• Adjustment range.
• Interface location.
• Strap pressure.
• Movement at maximum and minimum adjustment.
• Fit over clothing and PPE.
• Fit while standing, walking, bending and performing the real task.
• Ease of readjustment during a shift.
• Fit after repeated use, when straps and soft components settle.

Do not assume the research population represents your workforce

A 2026 systematic review analysed 191 arm- and back-support exoskeleton studies from 2019 to 2024. Participants were predominantly young and male, with approximately 78% male representation; none of the reviewed studies reported race or ethnicity, and anthropometric diversity was limited.[13]

This has direct buyer implications:

• Test the real workforce.
• Include women and workers across the size range.
• Do not approve a product after testing only one or two “average” users.
• Record who could not fit the device.
• Treat exclusion from the size range as a procurement outcome, not a user failure.

A device that performs well on one body type may not generalise to the full workforce.

—

6. Evaluate evidence by outcome, task and setting

Evidence hierarchy for buyers

From strongest to weakest:

1. Independent systematic reviews and meta-analyses.

2. Independent field studies in comparable real tasks.

3. Independent controlled laboratory studies.

4. Manufacturer-sponsored studies with transparent methods.

5. Internal pilot data from your own organisation.

6. Manufacturer demonstrations and testimonials.

7. Marketing claims without accessible methods or data.

A manufacturer study is not automatically invalid, but funding, methods, sample size, comparator and outcome selection should be disclosed.

Key evidence questions

• How many participants were tested?
• Were they representative of actual users?
• Was the study laboratory-based or conducted in the workplace?
• How long did the exposure last?
• Was there a comparison without the device?
• Was the order randomised?
• Were users familiarised with the device?
• Were both objective and subjective outcomes measured?
• Were adverse effects and task interference recorded?
• Were results reported by task?
• Was the study independent?
• Were statistical and practical significance both considered?

Laboratory benefit does not equal injury prevention

A reduction in electromyographic activity or joint moment can be useful, but it is a surrogate outcome. It does not by itself prove fewer injuries, fewer lost workdays or lower compensation costs.

The 2024 review of occupational exoskeleton studies found 49 eligible studies but emphasised the continued dominance of laboratory testing and the need for more real-world evaluation.[1]

A systematic review of quality, productivity and economic implications screened 6,722 records and included 15 relevant studies. None evaluated the economic implications of occupational exoskeleton use.[2]

Therefore, claims such as “this device will pay back in six months” require organisation-specific evidence and should not be inferred from short biomechanical studies.

—

7. Look for benefits and trade-offs at the same time

A good evaluation does not ask only whether support increases. It asks what changes elsewhere.

Possible positive outcomes

• Reduced activity in targeted muscles.
• Reduced joint moments or loading.
• Lower perceived exertion.
• Increased endurance for specific static tasks.
• Improved task tolerance.
• Better movement support.
• Increased confidence or mobility in appropriate clinical contexts.

Possible negative or compensatory outcomes

• Increased load in other muscles or joints.
• Restricted movement.
• Altered posture or technique.
• Contact pressure or discomfort.
• Heat accumulation.
• Balance effects.
• Interference with tools, PPE or surrounding structures.
• Slower movement.
• Reduced quality or dexterity.
• Difficulty sitting, climbing, crawling or entering vehicles.
• False confidence or overexertion.
• Reduced willingness to use the device.

A 2023 field study of a passive shoulder exoskeleton followed 10 logistics workers through a five-week familiarisation period. The device reduced anterior deltoid activity by 13% to 39% and upper trapezius activity by 16% to 60%, but the workers’ emotional response to using the device became less positive, raising questions about suitability for that environment.[14]

A 2026 aeronautics field study involving 10 male workers found 7% to 25% reductions in erector-spinae activity in four of seven polishing tasks, while acceptance was mixed but overall positive.[15]

These findings illustrate why procurement should not rely on one headline outcome.

—

8. Assess usability and acceptance as operational requirements

A technically effective device that workers avoid will not deliver sustained value.

Research on industrial exoskeleton acceptance repeatedly identifies comfort, task fit, perceived safety, usefulness and disruption as important factors.[16]

A field study of 146 logistics employees who used an active back exosuit for four hours found that 80% were willing to continue into a longer-term study. Intention to continue could be predicted with 78% accuracy using perceived effectiveness and work compatibility.[17]

This does not prove long-term adoption, but it shows that acceptance depends on whether users feel supported without unacceptable disruption.

Usability criteria

• Donning and doffing time.
• Number of adjustment steps.
• Ease of fitting without assistance.
• Comfort after 15 minutes, one hour and several hours.
• Heat and perspiration.
• Noise.
• Compatibility with communication and PPE.
• Movement restrictions.
• Ability to sit, walk, climb stairs and perform emergency movements.
• Cleaning requirements.
• Storage.
• Shift handover.
• Availability of multiple sizes.
• User confidence.
• Willingness to use the device voluntarily.

Familiarisation matters

Short demonstrations can overestimate or underestimate usability. Some initial discomfort may improve with adjustment and training, while other issues emerge only after repeated use. Buyers should plan repeated exposures before making a final decision.

—

9. Review safety and contextual risk

ASTM F3527 provides a task-specific risk-assessment framework and notes that risk assessment should involve people with knowledge of the task, the device and risk-analysis methods. It may require users, ergonomists, safety professionals, managers and manufacturer representatives.[6]

ASTM F3688 identifies ergonomic risk factors that can increase or decrease with exoskeleton use, including force, awkward posture, repetition, contact pressure, vibration and environmental stressors such as heat and cold.[18]

Minimum risk-review areas

• Pinch, crush and entanglement points.
• Emergency release.
• Battery, electrical and thermal risks.
• Unexpected movement or control failure.
• Falls and balance.
• Stairs, ladders and uneven surfaces.
• Interaction with vehicles and machinery.
• Confined spaces.
• Compatibility with fall protection.
• Fire and evacuation.
• Cleaning and cross-user hygiene.
• Skin pressure and irritation.
• User fatigue.
• Improper adjustment.
• Maintenance errors.
• Unauthorised modification.
• Data and cybersecurity for connected systems.

A risk assessment is not a one-time purchasing document. ASTM F3527 describes it as a living process that may need to be updated as tasks, users, equipment and work organisation change.[6]

—

10. Compare suppliers, not only devices

A technically promising device can fail operationally if the supplier cannot support deployment.

Supplier evaluation areas

Demand clarity on the exact configuration

Verify:

• Model and generation.
• Software version.
• Assistance setting.
• Size.
• Accessories.
• Battery.
• Harness and interface components.
• Tool or PPE integrations.

Performance data from an earlier model may not apply to the current configuration.

—

11. Calculate lifecycle cost, not only purchase price

The total cost can include:

• Purchase or lease.
• Shipping and import duties.
• Fitting.
• Training.
• Batteries and chargers.
• Software or data subscriptions.
• Maintenance.
• Spare parts.
• Cleaning.
• Storage.
• Downtime.
• Replacement components.
• Staff time for programme administration.
• Evaluation and occupational-health support.

Because published economic evidence remains weak, buyers should avoid using generic industry ROI calculators. Instead, build a local business case with clearly stated assumptions.

A defensible local business case

Compare:

1. Current task cost and exposure.

2. Cost of alternative controls.

3. Full exoskeleton lifecycle cost.

4. Measured pilot outcomes.

5. Adoption rate.

6. Operational disruption.

7. Uncertainty range.

8. Conditions required for scale.

Do not monetise injury prevention unless the relationship between the device, exposure reduction and injury outcome is supported well enough for your context.

—

12. Pilot before wider procurement

A pilot is not a product demonstration. It is a controlled decision process.

Minimum pilot stages

1. Define the problem.

2. Review higher-level controls.

3. Select candidate tasks.

4. Establish baseline measurements.

5. Define inclusion and exclusion criteria.

6. Shortlist devices.

7. Complete risk assessment.

8. Train users and supervisors.

9. Run repeated task trials.

1. Collect objective and subjective outcomes.
2. Record adverse effects and near misses.
3. Decide whether to reject, modify, extend or scale.

Suggested pilot outcomes

• Targeted muscle demand.
• Posture and movement.
• Perceived exertion.
• Local discomfort.
• Task quality.
• Task time.
• Error rate.
• Mobility.
• Balance.
• Heat.
• Fit.
• Donning time.
• Device faults.
• Worker acceptance.
• Supervisor observations.
• Maintenance burden.

Predefine stop conditions

Examples:

• Pain or numbness.
• Skin injury.
• Balance instability.
• Unsafe interaction with machinery.
• Emergency-egress interference.
• Device malfunction.
• Excessive heat.
• Unacceptable restriction.
• User request to stop.

The detailed pilot methodology should be documented before testing begins so that the decision is not shaped after seeing the results.

—

13. Exoskeleton Index buyer scorecard

The following framework is an editorial decision tool, not a regulatory standard. Buyers should adjust the weighting to their use case.

Suggested decision thresholds

• 80–100: Strong candidate for a controlled pilot or scale-up, subject to unresolved risks.
• 65–79: Potential candidate, but important conditions or evidence gaps remain.
• 50–64: Pilot only if the unmet need is significant and risks can be controlled.
• Below 50: Do not proceed without major changes or additional evidence.

A numeric score should never override a critical safety failure, regulatory problem or inability to fit intended users.

—

14. Supplier questions buyers should ask

Intended use

• What exact tasks and postures is this device designed for?
• Which tasks should it not be used for?
• What user groups are excluded?
• Is it intended for industrial, medical, rehabilitation, consumer or other use?

Performance

• What assistance is provided and how is it measured?
• Under which settings, loads and postures?
• Are results available for the current model?
• What independent field studies exist?
• What negative findings have been observed?

Fit and use

• What body dimensions are supported?
• How many sizes exist?
• How long does fitting take?
• Can users adjust it themselves?
• Is it compatible with our clothing, PPE and tools?
• What is the maximum recommended use time?

Safety

• What risk assessments have been completed?
• What are the emergency-release procedures?
• What happens after power loss or control failure?
• What fall, stumble, stairs or confined-space tests have been performed?
• What adverse events have been reported?

Operations

• What training is required?
• What maintenance is required?
• How are faults diagnosed?
• What service response time is available?
• What spare parts are stocked locally?
• What is the expected service life?

Commercial

• Purchase, lease or pilot price?
• What is included?
• What costs recur?
• What warranty applies?
• What insurance and liability arrangements exist?
• Can the supplier support a measured pilot?

—

15. Red flags

Be cautious when a supplier or article:

• Claims universal injury prevention.
• Guarantees productivity or ROI without transparent task-specific data.
• Uses only testimonials.
• Provides muscle-activity data without explaining the task.
• Does not disclose sample size.
• Uses research from a different model.
• Avoids discussing negative outcomes.
• Cannot provide fitting limits.
• Has no clear emergency-release procedure.
• Cannot explain local service and warranty.
• Presents CE marking as proof of effectiveness.
• Treats a medical, industrial and consumer claim as equivalent.
• Refuses a real-task pilot.
• Pressures users to continue despite discomfort.
• Collects data without clear privacy and security information.

—

Conclusion

The strongest buyer process is conservative, task-specific and evidence-aware.

An exoskeleton should not be selected because it looks advanced, has the highest assistance setting or appears in a successful demonstration. It should be selected because:

• The problem is clearly defined.
• Higher-level controls have been considered.
• The device matches the task and user population.
• Benefits and trade-offs are measured.
• Risks are understood and controlled.
• Workers can and will use it.
• The supplier can support it.
• The regulatory position is clear.
• A pilot demonstrates value in the actual context.

In many cases, the correct outcome of an evaluation may be to reject a device, narrow its use to one task, test a different category or redesign the work instead. That is a successful procurement process, not a failed one.

—

Methodology

This guide was prepared from official regulatory and standards information and peer-reviewed literature available up to 30 July 2026. Priority was given to official sources, systematic reviews and field studies. Manufacturer marketing claims were not used as independent evidence.

The guide is intended for market discovery and early-stage evaluation. It is not a substitute for medical, regulatory, occupational-safety, engineering or legal advice.

References

1. Cardoso A, Ribeiro A, Carneiro P, Colim A. Evaluating Exoskeletons for WMSD Prevention: A Systematic Review of Applications and Ergonomic Approach in Occupational Settings. International Journal of Environmental Research and Public Health. 2024. https://pubmed.ncbi.nlm.nih.gov/39767533/

2. Howard J, Murashov V, Lowe B, Lu J, et al. Quality, productivity, and economic implications of exoskeletons for occupational use: A systematic review. 2023. https://pubmed.ncbi.nlm.nih.gov/37368889/

3. National Institute for Occupational Safety and Health. Hierarchy of Controls. https://www.cdc.gov/niosh/hierarchy-of-controls/

4. International Organization for Standardization. ISO/CD 25563: Ergonomics — Process for the integration of wearable physical assistive devices. https://www.iso.org/standard/90736.html

5. ASTM International. Committee F48 on Exoskeletons and Exosuits. https://www.astm.org/membership-participation/technical-committees/committee-f48

6. ASTM International. F3527-24 Standard Guide for Assessing Risks Related to Implementation of Exoskeletons in Task-Specific Environments. https://store.astm.org/f3527-24.html

7. U.S. Food and Drug Administration. Powered lower extremity exoskeleton, product code PHL, 21 CFR 890.3480. https://www.accessdata.fda.gov/scripts/cdrh/cfdocs/cfpcd/classification.cfm?ID=PHL

8. European Union. Regulation (EU) 2017/745 on medical devices. https://eur-lex.europa.eu/legal-content/EN/TXT/?uri=celex%3A32017R0745

9. Golabchi A, Chao A, Tavakoli M. A Systematic Review of Industrial Exoskeletons for Injury Prevention: Efficacy Evaluation Metrics, Target Tasks, and Supported Body Postures. Sensors. 2022. https://pubmed.ncbi.nlm.nih.gov/35408328/

1. Kermavnar T, de Vries A, de Looze M, O’Sullivan L. Effects of industrial back-support exoskeletons on body loading and user experience: an updated systematic review. Ergonomics. 2021. https://pubmed.ncbi.nlm.nih.gov/33369518/
2. Schwartz M, et al. Biomechanical Consequences of Using Passive and Active Back-Support Exoskeletons during Different Manual Handling Tasks. 2023. https://pubmed.ncbi.nlm.nih.gov/37569010/
3. Li-Baboud Y, et al. Evaluation Methods and Measurement Challenges for Industrial Exoskeletons. Sensors. 2023. https://pubmed.ncbi.nlm.nih.gov/37420770/
4. Kazemi Z, Park J, Srinivasan D. Unequal assistance? A systematic review of demographic representation and outcome differences in exoskeleton studies. Applied Ergonomics. 2026. https://pubmed.ncbi.nlm.nih.gov/42114441/
5. Jakobsen LS, et al. Biomechanical changes, acceptance, and usability of a passive shoulder exoskeleton in manual material handling: A field study. Applied Ergonomics. 2023. https://pubmed.ncbi.nlm.nih.gov/37531933/
6. Field assessment of a passive back exoskeleton in the aeronautics industry: biomechanical responses and worker acceptance. 2026. https://pubmed.ncbi.nlm.nih.gov/42313610/
7. Elprama SA, Vanderborght B, Jacobs A. An industrial exoskeleton user acceptance framework based on a literature review of empirical studies. Applied Ergonomics. 2022. https://pubmed.ncbi.nlm.nih.gov/34847372/
8. Active back exosuits demonstrate positive usability perceptions that drive intention-to-use in the field among logistic warehouse workers. 2024. https://pubmed.ncbi.nlm.nih.gov/39388886/
9. ASTM International. F3688-25 Standard Guide for Evaluating Potential Ergonomic Risks From Exoskeleton Use. https://store.astm.org/f3688-25.html