Increasing Warehouse Productivity With an Ergonomic Handheld Scanner

Abstract

Rugged handheld scanners, known in the industry as rugged mobile computers, are used for critical warehouse operations, such as receiving, order picking, and put-away. The form of rugged scanners has not fundamentally changed since it was introduced to replace pen and clipboard. Warehouses have extracted the maximum available efficiency increases available through today’s handheld rugged scanners, but new operational challenges require new ways to further increase productivity and accuracy. The “line of sight” rugged handheld scanner concept described in this article is designed to enhance the user’s efficiency by eliminating non-value-added wrist motions.

Keywords

rugged handheld scanner “line-of-sight” display productivity usability warehouse operations supply chain order fulfillment

Increasing focus on speed and efficiency in order processing led to the design of a device that reduces musculoskeletal effort.

Order picking in a warehouse is a critical function in the supply chain to ensure accurate and timely delivery of product to the end customer. Therefore, picking throughput in the warehouse is dependent on the efficiency, accuracy, and productivity of the order picker (end user).

Over the past 20 years, warehouses have transitioned from paper-based picking systems to rugged handheld scanners. Warehouses have optimized these systems for their operations; however, they are still living with system inefficiencies as part of their cost of doing business. In the current world of e-commerce, multiple SKUs (stock-keeping units), and same-day delivery requirements from consumers, warehouses are looking for additional opportunities to increase productivity and accuracy.

This article provides an overview of the design, development, evaluation, and testing of a new rugged handheld scanner for the warehouse with a “line-of-sight” display configuration whereby the display is always in the user’s view during the data capture (scanning) process. The primary focus and purpose of the new design is to present critical information in a user-friendly manner, reduce non-value-added wrist motions and effort, and increase worker productivity for order picking.

Key Research Insights

A research study was initiated to investigate traditional rugged handheld scanner use in warehouses conducting e-commerce operations. Research was conducted in multiple warehouse locations across North America, Europe, and Asia. The research effort included interviews with multiple stakeholders, including operations supervisors and information technology managers; and observation and interaction with multiple warehouse workers performing picking. A key finding emerged from the research: The traditional “gun-style” rugged handheld scanners required order pickers to perform two steps to complete a “scan-and-verify” task: scan the barcode, and tilt the mobile device to verify the scanned information.

Figure 1 captures the “scan–verify” activity associated with a typical warehouse picking work flow.

Figure 1.

Typical warehouse scan–verify picking work flow.

Figure 2 shows a scan–verify picking operation.

Figure 2.

Scan–verify warehouse picking operation.

With an average of three sets of scan-and-verify actions per item picked, this non-value-added wrist motion interrupts the work flow and increases cycle time and motion overhead. The result is reduced worker productivity and efficiency, and increased worker fatigue.

Design Process

The design objective for the team was clear from the scan-and-verify research insights highlighted in the previous section: Eliminate wasted tilt motions and find untapped productivity gains.

Initial design premise

Multiple brainstorming sessions with the cross-functional team resulted in an initial idea of adding a secondary display to a traditional mobile computer form factor. This display would facilitate “information at a glance” (Figure 3).

Figure 3.

Initial concept for secondary display to facilitate information at a glance.

Additional design iterations resulted in the progressive refinement of this initial concept, eventually leading to a full reconfiguration of the concept with a vertically mounted display (Figure 4). This vertically mounted display was driven by the need to convey richer information, improve efficiency, and eliminate the wasted tilt motion.

Figure 4.

Full vertical display concept.

Concept design development phase

Once the full vertical display design premise was established, the lead designer started the design process with the human factors team. That team worked with the designer to evaluate concepts for grip sizing, display angle, and scan angle. Display sizing and access studies were conducted, and multiple heuristic evaluations were performed.

Early human factors testing (Phase 1 in the next section) showed that the new line-of-sight concept could replace the traditional dual-plane interface with a single plane that would enable users to scan and view the screen with a single motion. This line-of-sight display configuration should effectively eliminate the “tilt” motions completely and result in a major increase in worker productivity by reducing the physical motion and effort associated with picking. Figure 5 illustrates the idea.

Figure 5.

Single- versus dual-plane user interface.

Once the value of the new line-of-sight display was established through human factors validation (Phase 1), the business unit approved the project and the design process moved to the detail-design-and-implementation phase.

Engineering development phase

Design and human factors personnel led the effort throughout the implementation phase with engineering. Detailed design features and functionality (display orientation, grip features, weight, balance, scan angle, trigger actuation force and travel, optimized interaction with display and touch panel) were tested and implemented in this phase with a progression of prototype builds. Continuous human factors validation of the prototypes (including Phase 2 and Phase 3) was conducted during this phase, with the goal of quickly informing the design implementation phase.

Figure 6 illustrates the design process and design progression of the line-of-sight rugged handheld scanner from early concept to final design.

Figure 6.

Design process and evolution from concept to final design of line-of-sight rugged handheld scanner.

Phase 1: Early Mock-Up Testing

A representative mock-up was built for testing. The mock-up had a smartphone display aligned to the user’s line of sight, a scan engine positioned in the same plane as the display (like a smartphone camera), and gun grip handle with a trigger module (Figure 7).

Figure 7.

Early mock-up for human factors testing.

This mock-up was tested against a traditional gun-style mobile computer (Figure 8).

Figure 8.

Traditional gun-style mobile computer.

Goal

The goal was to investigate whether the new line-of-sight form factor might provide potential advantages when compared with the traditional gun-style form factor.

Users and tasks

Ten experienced warehouse workers (average experience = 10 years, average age = 34 years) executed 66 scans (two trials, 33 scans per trial) with each device in a simulated warehouse pick-and-pack operation at an internal testing facility. Users were asked to perform the following four picking task steps:

Scan the shelf location.

Verify the location code that was displayed on the device screen.

Scan the product at that location.

Verify the product number by looking at the information displayed on the screen.

The shelf heights tested are shown in Figure 9.

Figure 9.

Shelf height variation: Users wired with electromyography and goniometry.

Metrics measured

The following measurements were taken:

Muscle effort and wrist motions during scan–verify tasks

Efficiency of the device (time to task completion) measured and timed using a stopwatch

Perceived effort, comfort, accuracy, and ease of use (using Likert scales) administered post-task with each device

Subjective feedback of handle comfort, weight, balance, and preference administered posttest.

All sessions were videotaped.

Objective measurements

Users were wired to surface electromyography (SEMG) sensors and an electrogoniometer to capture and measure muscle effort and wrist motion metrics during execution of the scan-and-verify tasks (Kumar & Mital, 1996). The resulting data on the required muscle effort and postural deviations would indicate the physical demand on the user. The Noraxon Telemyo system with MyoResearch software was used for data acquisition and processing.

SEMG testing details

Muscle effort levels associated with the flexor carpi radialis (anterior forearm compartment) and extensor carpi ulnaris (posterior forearm compartment) muscles were monitored during the test.

For surface electrode attachment, the larger flexor carpi radialis and extensor carpi ulnaris muscles were located by having the participant flex and extend his or her wrist. The electrodes were attached after preparing the site (not shaved) with alcohol, and the resulting signals were checked on the Noraxon system to ensure that the identified longitudinal muscles closest to the surface were acting when the participant flexed and extended his or her wrist (Vanswearingen, 1983).

Each participant gripped and squeezed a Jamar grip dynamometer using a power grip to obtain a baseline maximum reference exertion (MRE) of the involved muscles (Seo, Armstrong, Ashton-Miller, & Chaffin, 2008). The corresponding SEMG signals were scaled using the MRE to obtain the percentage of muscle exertion associated with each subsequent test (%MRE). This normalization process enabled comparison of muscle effort between participants. All calibration was done in a neutral posture with the elbow joint supported and not in wrist flexion or extension (Konrad, 2005; Marras, 1992). The measured task was not highly dynamic and involved a series of slow scan–verify tasks at three shelf heights with the hand/forearm in a neutral prosupination posture with slight wrist extension. In the measured task, the wrist was mostly in radial/ulnar deviation.

For the type of task we were measuring, we felt that muscle activity recorded during a maximum power grip in a neutral wrist posture was sufficient for the comparisons we were interested in making (Duque, Masset, & Malchaire, 1995). We were not inferring any clinical implications from this study.

With a full grip dynamometer calibration in neutral posture, the key wrist movers for scanning were acting as required (verified by Noraxon muscle signals) and were a good reference contraction for the task we measured. The files were controlled, collected, and analyzed as single-period data files associated with the scan–verify tasks at each shelf height.

Impact of wrist postures

Hand strength is at its maximum when the hand and wrist are aligned in a neutral position with the wrist in slight extension. Any deviation results in a reduction of hand strength, which contributes to user fatigue, which in turn creates opportunity for error. For example, a radial deviation of 25° results in a 20% reduction in static hand strength, and an ulnar deviation of 40° results in a 25% reduction in static hand strength (Eastman Kodak Company, 1986; Plewa, Potvin, & Dickey, 2015; Snook, Vaillancourt, Ciriello, & Webster, 1997).

Gun-style mobile computers require multiple radial and ulnar deviations of the hand to tilt the device to scan and then view the data displayed on the screen. In contrast, the line-of-sight single-plane design potentially allows users to keep the hand, wrist, and forearm aligned in a more neutral posture while viewing the screen.

For this study, dynamic wrist postures of the dominant hand (the hand holding the device) were recorded using an electrogoniometer. Movements of the hand/wrist were measured in two anatomical planes: flexion/extension and ulnar/radial deviations.

Test results: Physical measurement metrics

The Phase 1 concept test revealed that productivity was increased by eliminating non-value-added wrist movements associated with scan-and-verify tasks during picking using a line-of-sight display configuration. Additionally, these gains were accomplished with less physical effort from users, enabling them to accomplish more tasks with less physical effort.

Table 1 shows the physical measurement metrics from the study. On average, the early line-of-sight mock-up showed a 23.1% reduction in wrist flexor muscle effort compared with the legacy handheld scanner. The early mock-up showed a 76.1% reduction in number of radial deviations compared with the legacy device during controlled scan-and-verify tasks at the three shelf heights (total of nine scan-and-verify tasks).

Table 1.

Physical Measurement Metrics (N = 10)

Measure	Legacy Device	Mock-Up
Average wrist flexor muscle effort (%MRE)	15.6 (3.3)	12.0 (2.6)
Average wrist extensor muscle effort (%MRE)	20.5 (4.2)	15.1 (3.1)
Mean number of radial deviations/pick location	9.2 (0.6)	2.2* (0.9)
Mean time to task completion (seconds)	181.0 (10.6)	147.0* (6.2)

Note. Standard errors in parentheses. %MRE = percentage of maximum reference exertion.

p ≤ .05.

Also, the early mock-up showed an 18.8% boost in productivity compared with the legacy handheld scanner at a scan rate of 14 items per minute.

Test results: Subjective ratings (comfort and product feature ratings)

Subjective questionnaires for body-part comfort, accuracy, and ease of use were administered upon task completion with each device using a 5-point Likert scale.

Figure 10 shows the Likert scale. Table 2 summarizes the posttask mean ratings for comfort, accuracy, and ease of use.

Figure 10.

Likert scale used.

Table 2.

Posttask Mean Ratings for Comfort, Accuracy, and Ease of Use (N = 10)

Variable	Legacy Device	Mock-Up
Comfort
Finger	4.4 (4.5)	4.7 (5)
Wrist	3.9 (4)	4.4 (5)
Hand	3.7 (4)	4.0 (5)
Forearm	3.6 (4)	4.7 (5)
Shoulder	3.8 (4)	4.7 (5)
Accuracy	4.3 (4)	4.9 (5)
Ease of use	4.2 (5)	4.9 (5)

Note. Medians in parentheses.

Additional Likert scale surveys were administered (posttest for all devices) for product features and functionality. Table 3 shows the mean posttest ratings for product features.

Table 3.

Posttest Mean Ratings for Product Features (N = 10)

Product Feature	Legacy Device	Mock-Up
Balance	2.9 (3)	4.5 (5)
Weight	2.5 (3)	4.3 (4.5)
Handle size	3.3 (3.5)	4.1 (5)
Efficiency for picking	3.6 (4)	5.0 (5)
Scan-and-verify effort	3.4 (4)	4.7 (5)
Overall	3.8 (4)	4.9 (5)

Note. Medians in parentheses.

Tables 2 and 3 show that the line-of-sight mock-up was subjectively rated higher compared with the legacy handheld scanner. Participants were also asked to pick their preferred device. Eight out of 10 users picked the line-of-sight mock-up as their preferred device.

The results from the Phase 1 early-concept human factors testing proved the value of the line-of-sight design and enabled the business unit to green-light the project for development.

Phase 2: Functioning Prototype Testing

The test protocol used for the functioning prototype was the same as Phase 1 testing with 10 experienced subjects (eight males, two females) using the same picking work flow used previously. Figure 11 shows the functioning prototype.

Figure 11.

Functioning prototype.

Figure 12 shows the picking tasks.

Figure 12.

Picking tasks.

The results of Phase 2 testing are shown in Table 4. The results in Table 4 align with the findings in Phase 1 testing of the early mock-up. For scan-and-verify tasks, the prototype showed a 15.3% reduction in wrist flexor muscle effort and a 34% reduction in wrist extensor muscle effort. The prototype also showed a 62.5% reduction in peak radial deviation for posture held at the end of scan-and-verify tasks.

Table 4.

Results of Phase 2 Testing (N = 10)

Measure	Legacy Device	Prototype
Average wrist flexor muscle effort (%MRE)	11.5	9.7
Average wrist extensor muscle effort (%MRE)	13.1	8.6
Peak radial deviation (degrees)^a	36.2	13.6*
Average time to task completion (seconds)	180.0	153.0*

Note. %MRE = percentage of maximum reference exertion.

Peak radial deviation refers to the aggregate of radial deviation data points at the end of the scan–look operation where the posture was held for approximately 1.25 s to verify the scan.

p ≤ .05.

Additionally, the prototype showed a 15% reduction in time to task completion.

Phase 3: Final Product Validation

The final testing of a functioning line-of-sight scanner versus gun-style scanner was conducted in a controlled testing environment in a warehouse conducted by a third-party human factors consultant.

Ten experienced warehouse pickers were recruited for the test. Objective metrics included muscle effort for scan-and-verify tasks, wrist deviations for scan-and-verify tasks, and time to overall task completion.

The test setting was a warehouse environment. Participants completed three picking trials with each device (25 pick locations per trial). They also completed a series of scan-and-verify tasks with each device at three shelf heights (five scan-and-verify sequences at each shelf height).

As before, metrics collected included the following:

Productivity (time to task completion)

Muscle effort (wrist flexors/extensors) and dynamic wrist posture

Subjective feedback of comfort, weight, balance, effort levels, and fatigue

Subjective ranking preference of device after completion of tasks

The testing order of devices was randomized, with the picking sequence constant. Muscle activity and wrist deviations were measured during the scan-and-verify tasks.

Table 5 summarizes the posttask mean ratings for body-part comfort. It shows that the final product was subjectively rated higher compared with the legacy handheld scanner in most comfort categories.

Table 5.

Posttask Mean Ratings for Body-Part Comfort (N = 10)

Comfort Category	Legacy Device	Final Product
Finger	4.4 (4)	4.4 (5)
Wrist	4.3 (4)	4.8 (5)
Hand	4.4 (4.5)	4.0 (4)
Forearm	4.2 (5)	4.6 (5)
Shoulder	4.7 (5)	4.9 (5)

Note. Medians in parentheses.

Table 6 shows the mean posttest ratings for product features. It shows that the final product was subjectively rated higher for overall use compared with the legacy handheld scanner. Lower mean ratings for speed and accuracy were related to the touch-only interface on the new device (one participant was uncomfortable with the touch-only interface). Participants were also asked to pick their preferred device. Six out of 10 users chose the final product as their preferred device, and four out of 10 picked the legacy device.

Table 6.

Posttest Mean Ratings for Product Features (N = 10)

Product Feature	Legacy Device	Final Product
Ease of use	4.6 (5)	4.6 (5)
Speed	4.7 (5)	4.4 (5)
Accuracy	4.7 (5)	4.3 (5)
Overall	4.2 (4)	4.7 (5)

Note. Medians in parentheses.

The results of the objective measurement metrics testing are shown in Table 7. As seen in that table, the line-of-sight scanner showed a reduction of 12.3% (high shelf) to 22.0% (low shelf) in muscle effort compared with the gun-style handheld scanner for scan and verify tasks. The line-of-sight scanner showed a 55.1% reduction in radial/ulnar wrist motion compared with the gun-style mobile handheld scanner across all shelf heights for scan-and-verify tasks.

Table 7.

Objective Metrics (N = 10)

Measure	Legacy Device	Final Product
Average muscle effort low shelf (%MRE)	24.1	18.8
Average muscle effort high shelf (%MRE)	20.4	17.9
Number total radial/ulnar wrist movements low shelf	173.0	69.8*
Number total radial/ulnar wrist movements middle shelf	116.0	59.6*
Number total radial/ulnar wrist movements high shelf	248.0	107.0*
Average time to pick task completion (seconds)	644.0	555.0*

Note. %MRE = percentage of maximum reference exertion.

p ≤ .05.

Additionally, the line-of-sight scanner showed a 13.8% reduction in time to task completion compared with the legacy handheld rugged scanner. These results aligned well with those from the previous phases of testing.

Conclusions

The final product validation test, along with the earlier concept and prototype tests, showed measurable productivity benefits for the line-of-sight rugged handheld scanner over traditional handheld rugged scanners. The line-of-sight rugged handheld scanner showed a 14% productivity increase over a traditional rugged handheld scanner.

Major productivity benefits were realized by eliminating wrist motions associated with scanning and verifying information on the display.

Footnotes

Chandra Nair is a human factors practitioner at Zebra Technologies, Inc., and a Certified Professional Ergonomist. He has extensive experience in testing and evaluation of products involving human interface, including handheld data capture devices, hand tools, surgical instruments, controls, and automotive and office seating. He can be reached at chandra.nair@zebra.com.

Konstantinos Tsiopanos is an industrial designer focused on advanced product design in the Innovation Design Team at Zebra Technologies. He has spent the past 6 years envisioning, designing, prototyping, and testing hardware and user experiences in the fields of enterprise-class laptop computing, public safety devices, barcode scanners, rugged handheld mobile computers, and mobile and desktop printers.

Richard Martin is a design researcher with the Innovation Design Team at Zebra Technologies. He spends most of his time gathering, validating, and championing user requirements in the field, in the studio, and in various conference rooms across the globe. In the past 10 years his work has covered retail, transport, logistics, warehouse, and public safety industries.

Graham Marshall has 35 years of experience in research, design, and development in both the United States and Canada. Throughout his career, he has led product innovation and human factors teams in a number of high-technology companies, including NCR, AT&T, Nortel, and Motorola. In addition, he has taught product design strategy and lectured on product innovation at conferences in North America, Europe, and Asia.

References

Duque

Masset

Malchaire

. (1995). Evaluation of handgrip force form EMG measurements. Applied Ergonomics, 26, 61–66.

Eastman Kodak Company. (1986). Ergonomic design for people at work (Vol. 2). Rochester, NY: Author.

Konrad

. (2005). The ABC of EMG: A practical introduction to kinesiological electromyography. Scottsdale, AZ: Noraxon U.S.A.

Kumar

Mital

. (1996). Electromyography in ergonomics. London, UK: Taylor & Francis.

Marras

. (1992). Applications of electromyography in ergonomics (chap. 7). DHHS (NIOSH) Publication No. 91-100. Morgantown, WV: U.S. Department of Health and Human Services, Public Health Service, Centers for Disease Control, National Institute for Occupational Safety and Health.

Plewa

Potvin

J. R.

Dickey

J. P

. (2015). Wrist rotations about one or two axes affect maximum wrist strength. Applied Ergonomics, 53(Part A), 152–160.

Seo

N. J.

Armstrong

T. J.

Ashton-Miller

J. A.

Chaffin

D. B

. (2008). Wrist strength is dependent on simultaneous power grip intensity. Ergonomics, 51, 1694–1605.

Snook

S. H.

Vaillancourt

D. R.

Ciriello

V. M.

Webster

B. S

. (1997). Maximum acceptable forces for repetitive ulnar deviation of the wrist. American Industrial Hygiene Association Journal, 58, 509–517.

Vanswearingen

J. M

. (1983). Measuring wrist muscle strength. Journal of Orthopaedic & Sports Physiological Therapy, 4, 217–228.