- Email: [email protected]
The influence of footwear tread groove parameters on available friction
Applied Ergonomics 50 (2015) 237e241
Contents lists available at ScienceDirect
Applied Ergonomics journal homepage: www.elsevier.com/locate/apergo
The influence of footwear tread groove parameters on available friction Mark G. Blanchette a, b, Christopher M. Powers b, * a b
Semper Scientific, Mission Viejo, CA, USA Division of Biokinesiology & Physical Therapy, University of Southern California, Los Angeles, CA, USA
a r t i c l e i n f o
a b s t r a c t
Article history: Received 7 December 2014 Accepted 26 March 2015 Available online
The purpose of this study was to determine how footwear tread groove parameters influence available friction (COF). Utilizing a whole shoe tester (SATRA STM 603), 3 groove parameters (width, depth and orientation) were evaluated. Groove orientation had 3 levels (parallel, oblique and perpendicular), width had 3 levels (3, 6 and 9 mm) and depth had 3 levels (2, 4 and 6 mm). In total, the COF of 27 shoes, each with a distinct groove combination, was assessed on wet porcelain tile. The 27 groove combinations produced a wide range of COF values (0.080e0.344). Groove orientation had the greatest impact on COF, explaining the greatest variance in observed COF values (s2 ¼ 0.81). The most slip resistant groove combination was an oblique orientation, with 3 mm width and 2 mm depth. The least slip resistant groove combination was a parallel orientation, with a 6 mm width and 6 mm depth. © 2015 Elsevier Ltd and The Ergonomics Society. All rights reserved.
Keywords: Slips Footwear Tread Friction
1. Introduction Slips are a common cause of falls in society. As such, there is a need to understand the factors that contribute to slip risk. Of the numerous factors that have been cited as contributing to slip risk, the one that is of great importance to the footwear industry is the shoe outsole. An outsole can be broken down into numerous components, 9 of which have been identified as possible contributors to available friction (COF): beveled edge, outsole material, outsole hardness, microscopic and macroscopic roughness, tread groove width, tread groove depth, tread groove orientation, and contact area (Chang, 2004; Chang et al., 2001a; Li and Chen, 2004; Li et al., 2006a; Li et al., 2006b; Wilson, 1990; Wilson, 1996). The function of outside tread is to facilitate ground contact, especially in the presence of a contaminant. Tread is important with respect to contaminant dispersion (Gronqvist, 1995), available friction (Wilson, 1990; Leclercq et al., 1995; Tisserand, 1985), and ultimately slip potential. Within the marketplace, outsole tread designs are highly variable with few visual consistencies. Currently, there are no
* Corresponding author. Division of Biokinesiology & Physical Therapy, University of Southern California, 1540 Alcazar St., CHP 155, Los Angeles, CA 90089, USA. Tel.: þ1 323 442 2948. E-mail address: [email protected] (C.M. Powers). http://dx.doi.org/10.1016/j.apergo.2015.03.018 0003-6870/© 2015 Elsevier Ltd and The Ergonomics Society. All rights reserved.
published guidelines on how to design outsole tread to meet specific levels of slip resistance. Given the complexity of the shoe-floor interface and the necessity for appropriate slip resistance, a scientific approach for outsole design is necessary. Several studies have investigated the influence of tread groove parameters on available friction. In 3 separate investigations, Li et al modified the test pad of an articulated strut tribometer to determine the influence of groove width, depth and orientation on COF (Li and Chen, 2004; Li et al., 2006a; Li et al., 2006b). In each study, available friction was assessed on 3 floor surfaces (vinyl, terrazzo and steel) with 3 different contaminants (water, detergent and oily). In their first study, Li and Chen (Li and Chen, 2004) evaluated the influence of groove width on COF using 5 test pad variations: smooth and 4 groove widths (3, 6, 9, and 12 mm). Each test pad had 3 equidistant parallel grooves set perpendicular to the direction of motion. The depth of all grooves was standardized to 7 mm. Results revealed that across the different floor surfaces, wider grooves resulted in greater COF for wet and water-detergent contaminants. In a second paper (Li et al., 2006b), the influence of groove depth on COF was determined with 10 test pad variations: 5 groove depths (1e5 mm in 1 mm increments) and 2 groove widths (3 mm and 9 mm). Results revealed that for wet and water-detergent conditions on all floor surfaces, COF increased with deeper and wider grooves. Lastly, the influence of groove orientation on COF was evaluated in a third paper (Li et al., 2006a). Using 6 test pad variations: 3 orientations (parallel, perpendicular and oblique to
238
M.G. Blanchette, C.M. Powers / Applied Ergonomics 50 (2015) 237e241
direction of motion) and 2 groove widths (3 mm and 9 mm), Li et al. reported that the 9 mm groove width had significantly greater COF values than the 3 mm width for both wet and water-detergent conditions. Furthermore, perpendicular and oblique groove orientations resulted in significantly greater COF values compared to parallel grooves. No difference in measured COF was found between the perpendicular and oblique groove orientations. Taken together, the studies of Li et al. (Li and Chen, 2004; Li et al., 2006a; Li et al., 2006b) suggest that wider, deeper, and perpendicular or oblique oriented tread grooves provide increased available friction in the presence of a water or water-detergent contaminant. However, a limitation of these investigations is that COF data were obtained using a portable tribometer that measured the available friction of a test pad made of a material not common to footwear outsoles (Neolite). Furthermore, the interaction of the 3 tread parameters (width, depth, orientation) was not evaluated. To date, no study has simultaneously evaluated the influence of all 3 tread groove parameters using actual shoes and with a common outsole material. The purpose of this study was to evaluate how various combinations of tread groove width, depth and orientation influence available friction as measured by the SATRA STM 603 whole shoe tester. More specifically, we sought to determine which tread groove parameter has the greatest impact on COF. A secondary aim was to determine what combination of width, depth and orientation results in the greatest slip resistance. 2. Methods 2.1. Procedures Twenty-seven pairs of men's size 10 shoes were assessed using a calibrated SATRA STM 603 whole shoe tester (SATRA, Kettering, Northamptonshire, UK; Fig. 1). Each shoe's tread differed in groove width, depth and orientation. Shoe outsoles were manufactured using 3 widths (3, 6 and 9 mm), 3 depths (2, 4 and 6 mm) and 3 orientations (parallel, perpendicular and oblique). Parallel grooves were directed along the length of the shoe. Perpendicular grooves were directed at a 90 angle to the length of the shoe. Oblique grooves were directed at a 45 angle to the length of the shoe (Fig. 2). Each shoe was constructed of the same upper and midsole combination. In addition, each outsole was constructed of the same material (styrene butadiene rubber) and hardness (62 Shore A) commonly used in the marketplace (Vibram USA, Concord, MA).
Fig. 2. Examples of study footwear. From left to right: parallel, oblique & perpendicular groove orientations. Each outsole depicted has 3 mm wide and 4 mm deep grooves.
Available friction testing followed the guidelines established in EN ISO 13287 (EN ISO 13287e2007, 2007) and the similar ASTM F2913 (ASTM F2913-11, 2011). The mechanical testing parameters outlined in both of these standards were adjusted as follows: normal force 400 N, sliding velocity 50 cm/s, and 9 shoe-floor contact angle. These parameters have been shown in a previous study to reduce COF assessment bias and improve slip prediction accuracy (Blanchette and Powers, 2015). All COF testing was performed on porcelain tile (ASTM ADJF250803 RS-B) contaminated with distilled water. This flooringcontaminant combination has previously been shown to produce slip events during walking (Blanchette and Powers, 2015; Powers et al., 2010). Five available friction measurements were obtained for each shoe and averaged for statistical analysis. 2.2. Statistical analysis To determine which groove parameter had the greatest impact on COF, an analysis of effect size was performed using the etasquared values obtained from a 3-way factorial ANOVA (width x depth x orientation) (Levine and Hullett, 2002). In total, 135 assessments of COF were included in the statistical analysis. To determine which combination of width, depth and orientation produced the greatest slip resistance, the COF of all 27 groove combinations was ranked from highest to lowest. All statistical analyses were performed using SPSS 18.0 Statistical Software (IBM Corporation, Armonk, NY). 3. Results
Fig. 1. The SATRA STM 603 whole shoe tester.
The 3-way ANOVA results and analysis of effect size are reported in Table 1. Overall, the 27 groove combinations produced a wide range of COF values (0.08e0.34; Table 2). The 3-way ANOVA revealed a significant 3-way interaction among groove width, depth and orientation (p < 0.001; Fig. 3). When collapsed across all levels of groove width and depth, groove orientation produced the greatest range of COF values (0.11e0.27; Table 3). The eta-squared analysis revealed that orientation had the greatest impact on COF, explaining 81% of the variance in COF. When collapsed across all levels of groove depth and orientation, groove width produced a small range of COF values (0.19e0.22; Table 3) and explained only 2% of the variance in COF. Similarly, when collapsed across all levels
M.G. Blanchette, C.M. Powers / Applied Ergonomics 50 (2015) 237e241 Table 1 ANOVA & effect size results for groove parameters on COF. Factor
F-value
p-Value
Effect size s2
Width Depth Orientation W*D W*O D*O W*D*O Error
58.5 98.9 2346.7 62.5 23.8 12.3 36.8
<0.001 <0.001 <0.001 <0.001 <0.001 <0.001 <0.001
0.02 0.03 0.81 0.04 0.02 0.01 0.05 0.02
W ¼ width; D ¼ depth; O ¼ orientation.
of groove width and orientation, groove depth produced a small range of COF values (0.19e0.22; Table 3) and explained only 3% of the variance in COF. All 27 tread groove combinations evaluated in the current study were ranked from highest to lowest based on COF (Table 2). There were 7 cases in which multiple tread groove combinations produced the same COF. In these cases, the tread groove combinations were given equivalent rankings. The most slip resistant groove combination was an oblique orientation, with a 3 mm width and 2 mm depth (0.34). The least slip resistant groove combination was a parallel orientation, with a 6 mm width and 6 mm depth (0.08). 4. Discussion The purpose of the current study was to determine the influence of tread groove width, depth and orientation on available friction. Our results indicated that of the 3 tread groove parameters, orientation had the greatest impact on COF, explaining 81% of the variance in measured COF. Interestingly, groove width and groove depth had a marginal impact on COF. Taken together, these 2 groove parameters only explained 5% of the variance in measured COF values.
Table 2 Rank of the average COF for each of the 27 tread groove combinations. Rank
COF
SD
Orientation
Width (mm)
Depth (mm)
1 2 3 6 6 6 7 8 10 10 12 12 14 14 15 17 17 18 19 20 22 22 24 24 25 26 27
0.34 0.31 0.30 0.28 0.28 0.28 0.26 0.25 0.24 0.24 0.23 0.23 0.22 0.22 0.21 0.20 0.20 0.19 0.16 0.14 0.12 0.12 0.11 0.11 0.10 0.09 0.08
0.01 0.01 0.01 0.01 0.02 0.01 0.02 0.01 0.01 0.02 0.01 0.01 0.01 0.01 0.02 0.00 0.01 0.01 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
OBL OBL OBL OBL PERP OBL PERP OBL PERP PERP OBL OBL OBL PERP PERP PERP PERP PERP PARA PARA PARA PARA PARA PARA PARA PARA PARA
3 9 6 3 3 3 6 9 6 9 6 9 6 9 3 3 9 6 9 3 3 6 3 9 6 9 6
2 4 2 6 2 4 4 2 6 2 4 6 6 4 6 4 6 2 4 2 6 2 4 2 4 6 6
SD ¼ Standard deviation, OBL ¼ Oblique; PERP ¼ Perpendicular; PARA ¼ parallel.
239
In a previous study, Li et al. (Li et al., 2006a) stated that parallel grooves were undesirable compared to oblique or perpendicular grooves in the presence of a contaminant. Based on the ranking of COF performance, our data support this claim. However, our results further suggest that oblique grooves may be preferred over perpendicular grooves. Oblique grooves may result in higher COF because of the orientation component in the direction of foot slide. This directional component may provide enhanced dynamic friction as an oblique orientation would provide resistance in the direction of foot slide while also dispersing water away from the shoe-floor interface. In contrast to the oblique orientation, perpendicular tread, provides high resistance in the direction of motion yet no directional dispersion of contaminant. This creates a “plowing” phenomenon at the shoe-floor interface. The plowing of contaminant between grooves would likely create a high frequency slip-stick response (Brown, 2006) due to the build-up of force against the leading edge of the groove and the build-up of contaminant within the groove, rather than the constant sliding friction provided by oblique grooves. Future research should focus on using human subjects to determine if the current results translate to slip potential and slip severity. The significant 3-way interaction among groove width, depth, and orientation indicated a complex relationship among the 3 tread groove parameters. While these complex interactions likely contributed to the lack of robust effects of width and depth on COF, several trends were apparent. First, the interaction between groove width and depth was consistent for both parallel and oblique groove orientations (Fig. 3). More specifically, COF was greatest for 3 mm width and lowest for 9 mm width when groove depth was 2 mm. When groove depth was 4 mm, COF was greatest for 9 mm width and lowest for 6 mm width. Additionally, when groove depth was 6 mm, COF was greatest for 3 mm width and lowest for 6 mm width. With respect to the perpendicular orientation however, when groove depth was 2 mm, COF was greatest for 3 mm width and lowest for 6 mm width. When groove depth was 4 mm, COF was greatest for 6 mm width and lowest for 3 mm width. Lastly, when groove depth was 6 mm, COF was greatest for 6 mm width and lowest for 9 mm width. Taken together, these trends suggest that when tread groove orientation has a component in the direction of foot slide, modification of groove width or depth may result in more predictable changes in available friction. The tread groove combination that measured the greatest COF was 3 mm wide, 2 mm deep and oblique orientation (0.34). A posthoc ManneWhitney U test indicated that this tread groove combination was ranked significantly higher than the 2nd ranked combination of 9 mm wide, 4 mm deep and oblique orientation (0.31; p ¼ 0.008). In contrast, the groove combination that measured the lowest COF was 6 mm wide, 6 mm deep and parallel orientation (0.08). A post-hoc ManneWhitney U test indicated that this tread groove combination was ranked significantly lower than the 26th ranked combination of 9 mm wide, 6 mm deep and parallel orientation (0.09; p ¼ 0.004). The highest performing groove combination and lowest performing groove combination differed in all 3 groove parameters. Although it was not possible to determine the exact cause of the observed differences in COF among groove combinations, a complex interaction likely existed among several factors that affected the rate of contaminant dispersion (i.e. width, depth, orientation, normal force, outsole material properties and hydrodynamic pressure). The rate of contaminant (e.g. distilled water) dispersion is a function of the squeeze-film effect, a determinant of available friction discussed by Li et al. and others (Chang et al., 2001a; Li and Chen, 2004; Li et al., 2006a; Li et al., 2006b; Chang et al., 2001b;
240
M.G. Blanchette, C.M. Powers / Applied Ergonomics 50 (2015) 237e241
Fig. 3. Graph of interactions among groove width, depth and orientation on COF.
Moore and Rayno, 1972). Further investigation is required to ascertain how groove parameters alter COF. Previous studies evaluating the influence of tread groove parameters on COF have utilized portable tribometers (Li and Chen, 2004; Li et al., 2006a; Li et al., 2006b). These studies revealed that groove width and depth had a robust influence on COF. Our study did not support this premise. One potential reason for this discrepancy is that the current study utilized a whole shoe tester. It is plausible that differences in the mechanical operating principles between tribometers and whole shoe testers could have contributed to the observed discrepancies between the current study and the investigations of Li et al. (Li and Chen, 2004; Li et al., 2006a; Li et al., 2006b). The Mark II tribometer used in the studies of Li et al. (Li and Chen, 2004; Li et al., 2006a; Li et al., 2006b) is an articulated-strut/inclined-strut class tribometer, which applies a 44 N gravity driven load to a Neolite test pad-floor interface. In contrast, the SATRA STM 603 is a whole shoe tester that uses a stationary shoe last with a pneumatically driven 400 N normal load. The floor slides underneath the shoe at a constant velocity and a force transducer measures the shear force during foot slide. Differences in mechanical operating principles between the Mark II and SATRA STM 603 (variable angle vs. constant angle, variable load vs. constant load and contact velocity) could affect the interactions at the sliding interface, thereby directly influencing the effects of groove parameters on available friction and thus reported COF values. Another explanation for the observed differences in groove parameter effects between the current study and previous investigations may be related to the use of different test feet. In the current study, the SATRA STM 603 used shoe outsoles comprised of 62 A Shore hardness styrene butadiene rubber (SBR). The Mark II tribometer used in the studies of Li et al. (Li and Chen, 2004; Li et al., 2006a; Li et al., 2006b) employed a square Neolite test pad (Shore A hardness 88e96). We suspect that the interactions occurring at the sliding interface are markedly different between these 2 testing devices. More specifically, the square and rigid test pad of the Mark II is unlikely to behave like the softer, flexible and contoured heel of a shoe under tribological conditions. Furthermore, we postulate that under the higher loads experienced with SATRA STM 603
testing, the softer SBR outsole material may affect the inherent groove structure through deformation. There are several limitations of the current study that need to be acknowledged. First, use of the SATRA STM 603 does not allow for detailed examination and analysis of the interactions occurring at the shoe-floor interface during foot slide (e.g. monitoring for changes in groove dimensions). The ability to do so could provide important information on why the groove effects observed in the current study differed from previous investigations. A second limitation is that only one shoe sole material and hardness combination was evaluated. As such, findings may not be generalizeable to other shoe types and outsole materials. In the current study, the shoe type was a casual walking shoe with a medium hardness SBR outsole. It is plausible that different groove effects would be observed with the use of alternative shoe types (e.g. running or dress shoes) and outsole materials (e.g. EVA, TPU, leather, etc.). 5. Conclusion The primary finding of this study was that of the 3 tread groove parameters (width, depth, and orientation) assessed with the SATRA STM 603, orientation had the greatest impact on available friction. These results indicate that groove orientation is an important component for available friction and should be considered when designing footwear outsoles for slip resistance. Future research should focus on tread groove orientation affects slip potential using human subjects. Acknowledgments Graduate student support provided by Division of Biokinesiology & Physical Therapy at USC and the ASTM International Graduate Student Scholarship. The authors of this paper would like to thank Vibram USA and the Quabaug Corporation for providing the footwear used in this study. Additionally, we would like to acknowledge the Oakley R&D Performance Lab for permitting access to their SATRA STM 603. References
Table 3 Mean COF data for each level of the 3 tread groove parameters. Parameter
Levels
Orientation COF
Parallel 0.11
Perpendicular 0.23
Oblique 0.27
Width COF
3 mm 0.22
6 mm 0.19
9 mm 0.20
Depth COF
2 mm 0.22
4 mm 0.21
6 mm 0.19
ASTM F2913-11, 2011. Standard Test Method for Measuring the Coefficient of Friction for Evaluation of Slip Performance of Footwear and Test Surfaces/Flooring Using a Whole Shoe Tester. ASTM International, West Conshohocken, PA. Blanchette, M.G., Powers, C.M., 2015. Slip prediction accuracy and bias of the SATRA STM 603 whole shoe tester. J. Test. Eval. (3), 43. Brown, R., 2006. Physical Testing of Rubber, fourth ed. Springer Science & Business Media Inc., New York, NY. 387. Chang, W.R., 2004. Preferred surface microscopic geometric features on floors as potential interventions for slip and fall accidents on liquid contaminated surfaces. J. Saf. Res. 35, 71e79. Chang, W.R., Gronqvist, R., Leclercq, S., 2001a. The role of friction in the measurement of slipperiness, part 1: friction mechanisms and definition of test conditions. Ergonomics 44 (13), 1217e1232.
M.G. Blanchette, C.M. Powers / Applied Ergonomics 50 (2015) 237e241 Chang, W.R., et al., 2001b. The role of friction in the measurement of slipperiness, part 2: survey of friction measurement devices. Ergonomics 44 (13), 1233e1261. EN ISO 13287e2007, 2007. Personal Protective Equipment. Footwear. Test Method for Slip Resistance ISO. Geneva, Switzerland. Gronqvist, R., 1995. Mechanisms of friction and assessment of slip resistance of new and used footwear soles on contaminated floors. Ergonomics 38 (2), 224e241. Leclercq, S., Tisserand, M., Saulnier, H., 1995. Tribological concepts involved in slipping accident analysis. Ergonomics 38 (2), 197e208. Levine, T.R., Hullett, C.R., 2002. Eta squared, partial eta squared, and misreporting of effect size in communication research. Hum. Commun. Res. 28 (4), 612e625. Li, K.W., Chen, C.J., 2004. The effect of shoe soling tread groove width on the coefficient of friction with different sole materials, floors, and contaminants. Appl. Ergon. 35, 499e507. Li, K.W., Chen, C.J., Lin, Ching-Hua, et al., 2006a. Relationship between measured friction coefficients and two tread groove design parameters for footwear pads. Tsinghua Sci. Technol. 11 (6), 712e719.
241
Li, K.W., Wu, H.H., Lin, Y.C., 2006b. The effect of shoe sole tread groove depth on the friction coefficient with different tread groove widths, floors and contaminants. Appl. Ergon. 37, 743e748. Moore, D.F., 1972. The Friction and lubrication of elastomers. In: Rayno, G.V. (Ed.), International Series of Monographs on Material Science and Technology. Pergamon Press, Oxford. Powers, C.M., Blanchette, M.G., Brault, J.R., et al., 2010. Validation of walkway tribometers: establishing a reference standard. J. Forensic Sci. 55 (2), 366e370. Tisserand, M., 1985. Progress in the prevention of falls caused by slipping. Ergonomics 28 (7), 1027e1042. Wilson, M., 1996. Slip resistance characteristics of footwear solings assessed using the SATRA friction tester. J. Test. Eval. 24 (6), 377e385. Wilson, M., 1990. Development of SATRA slip test and tread pattern design guidelines. In: Gray, B.E. (Ed.), Slips, Stumbles, and Falls: Pedestrian Footwear and Surfaces. American Society of Testing and Materials, Philadelphia, pp. 113e123.
Contents lists available at ScienceDirect
Applied Ergonomics journal homepage: www.elsevier.com/locate/apergo
The influence of footwear tread groove parameters on available friction Mark G. Blanchette a, b, Christopher M. Powers b, * a b
Semper Scientific, Mission Viejo, CA, USA Division of Biokinesiology & Physical Therapy, University of Southern California, Los Angeles, CA, USA
a r t i c l e i n f o
a b s t r a c t
Article history: Received 7 December 2014 Accepted 26 March 2015 Available online
The purpose of this study was to determine how footwear tread groove parameters influence available friction (COF). Utilizing a whole shoe tester (SATRA STM 603), 3 groove parameters (width, depth and orientation) were evaluated. Groove orientation had 3 levels (parallel, oblique and perpendicular), width had 3 levels (3, 6 and 9 mm) and depth had 3 levels (2, 4 and 6 mm). In total, the COF of 27 shoes, each with a distinct groove combination, was assessed on wet porcelain tile. The 27 groove combinations produced a wide range of COF values (0.080e0.344). Groove orientation had the greatest impact on COF, explaining the greatest variance in observed COF values (s2 ¼ 0.81). The most slip resistant groove combination was an oblique orientation, with 3 mm width and 2 mm depth. The least slip resistant groove combination was a parallel orientation, with a 6 mm width and 6 mm depth. © 2015 Elsevier Ltd and The Ergonomics Society. All rights reserved.
Keywords: Slips Footwear Tread Friction
1. Introduction Slips are a common cause of falls in society. As such, there is a need to understand the factors that contribute to slip risk. Of the numerous factors that have been cited as contributing to slip risk, the one that is of great importance to the footwear industry is the shoe outsole. An outsole can be broken down into numerous components, 9 of which have been identified as possible contributors to available friction (COF): beveled edge, outsole material, outsole hardness, microscopic and macroscopic roughness, tread groove width, tread groove depth, tread groove orientation, and contact area (Chang, 2004; Chang et al., 2001a; Li and Chen, 2004; Li et al., 2006a; Li et al., 2006b; Wilson, 1990; Wilson, 1996). The function of outside tread is to facilitate ground contact, especially in the presence of a contaminant. Tread is important with respect to contaminant dispersion (Gronqvist, 1995), available friction (Wilson, 1990; Leclercq et al., 1995; Tisserand, 1985), and ultimately slip potential. Within the marketplace, outsole tread designs are highly variable with few visual consistencies. Currently, there are no
* Corresponding author. Division of Biokinesiology & Physical Therapy, University of Southern California, 1540 Alcazar St., CHP 155, Los Angeles, CA 90089, USA. Tel.: þ1 323 442 2948. E-mail address: [email protected] (C.M. Powers). http://dx.doi.org/10.1016/j.apergo.2015.03.018 0003-6870/© 2015 Elsevier Ltd and The Ergonomics Society. All rights reserved.
published guidelines on how to design outsole tread to meet specific levels of slip resistance. Given the complexity of the shoe-floor interface and the necessity for appropriate slip resistance, a scientific approach for outsole design is necessary. Several studies have investigated the influence of tread groove parameters on available friction. In 3 separate investigations, Li et al modified the test pad of an articulated strut tribometer to determine the influence of groove width, depth and orientation on COF (Li and Chen, 2004; Li et al., 2006a; Li et al., 2006b). In each study, available friction was assessed on 3 floor surfaces (vinyl, terrazzo and steel) with 3 different contaminants (water, detergent and oily). In their first study, Li and Chen (Li and Chen, 2004) evaluated the influence of groove width on COF using 5 test pad variations: smooth and 4 groove widths (3, 6, 9, and 12 mm). Each test pad had 3 equidistant parallel grooves set perpendicular to the direction of motion. The depth of all grooves was standardized to 7 mm. Results revealed that across the different floor surfaces, wider grooves resulted in greater COF for wet and water-detergent contaminants. In a second paper (Li et al., 2006b), the influence of groove depth on COF was determined with 10 test pad variations: 5 groove depths (1e5 mm in 1 mm increments) and 2 groove widths (3 mm and 9 mm). Results revealed that for wet and water-detergent conditions on all floor surfaces, COF increased with deeper and wider grooves. Lastly, the influence of groove orientation on COF was evaluated in a third paper (Li et al., 2006a). Using 6 test pad variations: 3 orientations (parallel, perpendicular and oblique to
238
M.G. Blanchette, C.M. Powers / Applied Ergonomics 50 (2015) 237e241
direction of motion) and 2 groove widths (3 mm and 9 mm), Li et al. reported that the 9 mm groove width had significantly greater COF values than the 3 mm width for both wet and water-detergent conditions. Furthermore, perpendicular and oblique groove orientations resulted in significantly greater COF values compared to parallel grooves. No difference in measured COF was found between the perpendicular and oblique groove orientations. Taken together, the studies of Li et al. (Li and Chen, 2004; Li et al., 2006a; Li et al., 2006b) suggest that wider, deeper, and perpendicular or oblique oriented tread grooves provide increased available friction in the presence of a water or water-detergent contaminant. However, a limitation of these investigations is that COF data were obtained using a portable tribometer that measured the available friction of a test pad made of a material not common to footwear outsoles (Neolite). Furthermore, the interaction of the 3 tread parameters (width, depth, orientation) was not evaluated. To date, no study has simultaneously evaluated the influence of all 3 tread groove parameters using actual shoes and with a common outsole material. The purpose of this study was to evaluate how various combinations of tread groove width, depth and orientation influence available friction as measured by the SATRA STM 603 whole shoe tester. More specifically, we sought to determine which tread groove parameter has the greatest impact on COF. A secondary aim was to determine what combination of width, depth and orientation results in the greatest slip resistance. 2. Methods 2.1. Procedures Twenty-seven pairs of men's size 10 shoes were assessed using a calibrated SATRA STM 603 whole shoe tester (SATRA, Kettering, Northamptonshire, UK; Fig. 1). Each shoe's tread differed in groove width, depth and orientation. Shoe outsoles were manufactured using 3 widths (3, 6 and 9 mm), 3 depths (2, 4 and 6 mm) and 3 orientations (parallel, perpendicular and oblique). Parallel grooves were directed along the length of the shoe. Perpendicular grooves were directed at a 90 angle to the length of the shoe. Oblique grooves were directed at a 45 angle to the length of the shoe (Fig. 2). Each shoe was constructed of the same upper and midsole combination. In addition, each outsole was constructed of the same material (styrene butadiene rubber) and hardness (62 Shore A) commonly used in the marketplace (Vibram USA, Concord, MA).
Fig. 2. Examples of study footwear. From left to right: parallel, oblique & perpendicular groove orientations. Each outsole depicted has 3 mm wide and 4 mm deep grooves.
Available friction testing followed the guidelines established in EN ISO 13287 (EN ISO 13287e2007, 2007) and the similar ASTM F2913 (ASTM F2913-11, 2011). The mechanical testing parameters outlined in both of these standards were adjusted as follows: normal force 400 N, sliding velocity 50 cm/s, and 9 shoe-floor contact angle. These parameters have been shown in a previous study to reduce COF assessment bias and improve slip prediction accuracy (Blanchette and Powers, 2015). All COF testing was performed on porcelain tile (ASTM ADJF250803 RS-B) contaminated with distilled water. This flooringcontaminant combination has previously been shown to produce slip events during walking (Blanchette and Powers, 2015; Powers et al., 2010). Five available friction measurements were obtained for each shoe and averaged for statistical analysis. 2.2. Statistical analysis To determine which groove parameter had the greatest impact on COF, an analysis of effect size was performed using the etasquared values obtained from a 3-way factorial ANOVA (width x depth x orientation) (Levine and Hullett, 2002). In total, 135 assessments of COF were included in the statistical analysis. To determine which combination of width, depth and orientation produced the greatest slip resistance, the COF of all 27 groove combinations was ranked from highest to lowest. All statistical analyses were performed using SPSS 18.0 Statistical Software (IBM Corporation, Armonk, NY). 3. Results
Fig. 1. The SATRA STM 603 whole shoe tester.
The 3-way ANOVA results and analysis of effect size are reported in Table 1. Overall, the 27 groove combinations produced a wide range of COF values (0.08e0.34; Table 2). The 3-way ANOVA revealed a significant 3-way interaction among groove width, depth and orientation (p < 0.001; Fig. 3). When collapsed across all levels of groove width and depth, groove orientation produced the greatest range of COF values (0.11e0.27; Table 3). The eta-squared analysis revealed that orientation had the greatest impact on COF, explaining 81% of the variance in COF. When collapsed across all levels of groove depth and orientation, groove width produced a small range of COF values (0.19e0.22; Table 3) and explained only 2% of the variance in COF. Similarly, when collapsed across all levels
M.G. Blanchette, C.M. Powers / Applied Ergonomics 50 (2015) 237e241 Table 1 ANOVA & effect size results for groove parameters on COF. Factor
F-value
p-Value
Effect size s2
Width Depth Orientation W*D W*O D*O W*D*O Error
58.5 98.9 2346.7 62.5 23.8 12.3 36.8
<0.001 <0.001 <0.001 <0.001 <0.001 <0.001 <0.001
0.02 0.03 0.81 0.04 0.02 0.01 0.05 0.02
W ¼ width; D ¼ depth; O ¼ orientation.
of groove width and orientation, groove depth produced a small range of COF values (0.19e0.22; Table 3) and explained only 3% of the variance in COF. All 27 tread groove combinations evaluated in the current study were ranked from highest to lowest based on COF (Table 2). There were 7 cases in which multiple tread groove combinations produced the same COF. In these cases, the tread groove combinations were given equivalent rankings. The most slip resistant groove combination was an oblique orientation, with a 3 mm width and 2 mm depth (0.34). The least slip resistant groove combination was a parallel orientation, with a 6 mm width and 6 mm depth (0.08). 4. Discussion The purpose of the current study was to determine the influence of tread groove width, depth and orientation on available friction. Our results indicated that of the 3 tread groove parameters, orientation had the greatest impact on COF, explaining 81% of the variance in measured COF. Interestingly, groove width and groove depth had a marginal impact on COF. Taken together, these 2 groove parameters only explained 5% of the variance in measured COF values.
Table 2 Rank of the average COF for each of the 27 tread groove combinations. Rank
COF
SD
Orientation
Width (mm)
Depth (mm)
1 2 3 6 6 6 7 8 10 10 12 12 14 14 15 17 17 18 19 20 22 22 24 24 25 26 27
0.34 0.31 0.30 0.28 0.28 0.28 0.26 0.25 0.24 0.24 0.23 0.23 0.22 0.22 0.21 0.20 0.20 0.19 0.16 0.14 0.12 0.12 0.11 0.11 0.10 0.09 0.08
0.01 0.01 0.01 0.01 0.02 0.01 0.02 0.01 0.01 0.02 0.01 0.01 0.01 0.01 0.02 0.00 0.01 0.01 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
OBL OBL OBL OBL PERP OBL PERP OBL PERP PERP OBL OBL OBL PERP PERP PERP PERP PERP PARA PARA PARA PARA PARA PARA PARA PARA PARA
3 9 6 3 3 3 6 9 6 9 6 9 6 9 3 3 9 6 9 3 3 6 3 9 6 9 6
2 4 2 6 2 4 4 2 6 2 4 6 6 4 6 4 6 2 4 2 6 2 4 2 4 6 6
SD ¼ Standard deviation, OBL ¼ Oblique; PERP ¼ Perpendicular; PARA ¼ parallel.
239
In a previous study, Li et al. (Li et al., 2006a) stated that parallel grooves were undesirable compared to oblique or perpendicular grooves in the presence of a contaminant. Based on the ranking of COF performance, our data support this claim. However, our results further suggest that oblique grooves may be preferred over perpendicular grooves. Oblique grooves may result in higher COF because of the orientation component in the direction of foot slide. This directional component may provide enhanced dynamic friction as an oblique orientation would provide resistance in the direction of foot slide while also dispersing water away from the shoe-floor interface. In contrast to the oblique orientation, perpendicular tread, provides high resistance in the direction of motion yet no directional dispersion of contaminant. This creates a “plowing” phenomenon at the shoe-floor interface. The plowing of contaminant between grooves would likely create a high frequency slip-stick response (Brown, 2006) due to the build-up of force against the leading edge of the groove and the build-up of contaminant within the groove, rather than the constant sliding friction provided by oblique grooves. Future research should focus on using human subjects to determine if the current results translate to slip potential and slip severity. The significant 3-way interaction among groove width, depth, and orientation indicated a complex relationship among the 3 tread groove parameters. While these complex interactions likely contributed to the lack of robust effects of width and depth on COF, several trends were apparent. First, the interaction between groove width and depth was consistent for both parallel and oblique groove orientations (Fig. 3). More specifically, COF was greatest for 3 mm width and lowest for 9 mm width when groove depth was 2 mm. When groove depth was 4 mm, COF was greatest for 9 mm width and lowest for 6 mm width. Additionally, when groove depth was 6 mm, COF was greatest for 3 mm width and lowest for 6 mm width. With respect to the perpendicular orientation however, when groove depth was 2 mm, COF was greatest for 3 mm width and lowest for 6 mm width. When groove depth was 4 mm, COF was greatest for 6 mm width and lowest for 3 mm width. Lastly, when groove depth was 6 mm, COF was greatest for 6 mm width and lowest for 9 mm width. Taken together, these trends suggest that when tread groove orientation has a component in the direction of foot slide, modification of groove width or depth may result in more predictable changes in available friction. The tread groove combination that measured the greatest COF was 3 mm wide, 2 mm deep and oblique orientation (0.34). A posthoc ManneWhitney U test indicated that this tread groove combination was ranked significantly higher than the 2nd ranked combination of 9 mm wide, 4 mm deep and oblique orientation (0.31; p ¼ 0.008). In contrast, the groove combination that measured the lowest COF was 6 mm wide, 6 mm deep and parallel orientation (0.08). A post-hoc ManneWhitney U test indicated that this tread groove combination was ranked significantly lower than the 26th ranked combination of 9 mm wide, 6 mm deep and parallel orientation (0.09; p ¼ 0.004). The highest performing groove combination and lowest performing groove combination differed in all 3 groove parameters. Although it was not possible to determine the exact cause of the observed differences in COF among groove combinations, a complex interaction likely existed among several factors that affected the rate of contaminant dispersion (i.e. width, depth, orientation, normal force, outsole material properties and hydrodynamic pressure). The rate of contaminant (e.g. distilled water) dispersion is a function of the squeeze-film effect, a determinant of available friction discussed by Li et al. and others (Chang et al., 2001a; Li and Chen, 2004; Li et al., 2006a; Li et al., 2006b; Chang et al., 2001b;
240
M.G. Blanchette, C.M. Powers / Applied Ergonomics 50 (2015) 237e241
Fig. 3. Graph of interactions among groove width, depth and orientation on COF.
Moore and Rayno, 1972). Further investigation is required to ascertain how groove parameters alter COF. Previous studies evaluating the influence of tread groove parameters on COF have utilized portable tribometers (Li and Chen, 2004; Li et al., 2006a; Li et al., 2006b). These studies revealed that groove width and depth had a robust influence on COF. Our study did not support this premise. One potential reason for this discrepancy is that the current study utilized a whole shoe tester. It is plausible that differences in the mechanical operating principles between tribometers and whole shoe testers could have contributed to the observed discrepancies between the current study and the investigations of Li et al. (Li and Chen, 2004; Li et al., 2006a; Li et al., 2006b). The Mark II tribometer used in the studies of Li et al. (Li and Chen, 2004; Li et al., 2006a; Li et al., 2006b) is an articulated-strut/inclined-strut class tribometer, which applies a 44 N gravity driven load to a Neolite test pad-floor interface. In contrast, the SATRA STM 603 is a whole shoe tester that uses a stationary shoe last with a pneumatically driven 400 N normal load. The floor slides underneath the shoe at a constant velocity and a force transducer measures the shear force during foot slide. Differences in mechanical operating principles between the Mark II and SATRA STM 603 (variable angle vs. constant angle, variable load vs. constant load and contact velocity) could affect the interactions at the sliding interface, thereby directly influencing the effects of groove parameters on available friction and thus reported COF values. Another explanation for the observed differences in groove parameter effects between the current study and previous investigations may be related to the use of different test feet. In the current study, the SATRA STM 603 used shoe outsoles comprised of 62 A Shore hardness styrene butadiene rubber (SBR). The Mark II tribometer used in the studies of Li et al. (Li and Chen, 2004; Li et al., 2006a; Li et al., 2006b) employed a square Neolite test pad (Shore A hardness 88e96). We suspect that the interactions occurring at the sliding interface are markedly different between these 2 testing devices. More specifically, the square and rigid test pad of the Mark II is unlikely to behave like the softer, flexible and contoured heel of a shoe under tribological conditions. Furthermore, we postulate that under the higher loads experienced with SATRA STM 603
testing, the softer SBR outsole material may affect the inherent groove structure through deformation. There are several limitations of the current study that need to be acknowledged. First, use of the SATRA STM 603 does not allow for detailed examination and analysis of the interactions occurring at the shoe-floor interface during foot slide (e.g. monitoring for changes in groove dimensions). The ability to do so could provide important information on why the groove effects observed in the current study differed from previous investigations. A second limitation is that only one shoe sole material and hardness combination was evaluated. As such, findings may not be generalizeable to other shoe types and outsole materials. In the current study, the shoe type was a casual walking shoe with a medium hardness SBR outsole. It is plausible that different groove effects would be observed with the use of alternative shoe types (e.g. running or dress shoes) and outsole materials (e.g. EVA, TPU, leather, etc.). 5. Conclusion The primary finding of this study was that of the 3 tread groove parameters (width, depth, and orientation) assessed with the SATRA STM 603, orientation had the greatest impact on available friction. These results indicate that groove orientation is an important component for available friction and should be considered when designing footwear outsoles for slip resistance. Future research should focus on tread groove orientation affects slip potential using human subjects. Acknowledgments Graduate student support provided by Division of Biokinesiology & Physical Therapy at USC and the ASTM International Graduate Student Scholarship. The authors of this paper would like to thank Vibram USA and the Quabaug Corporation for providing the footwear used in this study. Additionally, we would like to acknowledge the Oakley R&D Performance Lab for permitting access to their SATRA STM 603. References
Table 3 Mean COF data for each level of the 3 tread groove parameters. Parameter
Levels
Orientation COF
Parallel 0.11
Perpendicular 0.23
Oblique 0.27
Width COF
3 mm 0.22
6 mm 0.19
9 mm 0.20
Depth COF
2 mm 0.22
4 mm 0.21
6 mm 0.19
ASTM F2913-11, 2011. Standard Test Method for Measuring the Coefficient of Friction for Evaluation of Slip Performance of Footwear and Test Surfaces/Flooring Using a Whole Shoe Tester. ASTM International, West Conshohocken, PA. Blanchette, M.G., Powers, C.M., 2015. Slip prediction accuracy and bias of the SATRA STM 603 whole shoe tester. J. Test. Eval. (3), 43. Brown, R., 2006. Physical Testing of Rubber, fourth ed. Springer Science & Business Media Inc., New York, NY. 387. Chang, W.R., 2004. Preferred surface microscopic geometric features on floors as potential interventions for slip and fall accidents on liquid contaminated surfaces. J. Saf. Res. 35, 71e79. Chang, W.R., Gronqvist, R., Leclercq, S., 2001a. The role of friction in the measurement of slipperiness, part 1: friction mechanisms and definition of test conditions. Ergonomics 44 (13), 1217e1232.
M.G. Blanchette, C.M. Powers / Applied Ergonomics 50 (2015) 237e241 Chang, W.R., et al., 2001b. The role of friction in the measurement of slipperiness, part 2: survey of friction measurement devices. Ergonomics 44 (13), 1233e1261. EN ISO 13287e2007, 2007. Personal Protective Equipment. Footwear. Test Method for Slip Resistance ISO. Geneva, Switzerland. Gronqvist, R., 1995. Mechanisms of friction and assessment of slip resistance of new and used footwear soles on contaminated floors. Ergonomics 38 (2), 224e241. Leclercq, S., Tisserand, M., Saulnier, H., 1995. Tribological concepts involved in slipping accident analysis. Ergonomics 38 (2), 197e208. Levine, T.R., Hullett, C.R., 2002. Eta squared, partial eta squared, and misreporting of effect size in communication research. Hum. Commun. Res. 28 (4), 612e625. Li, K.W., Chen, C.J., 2004. The effect of shoe soling tread groove width on the coefficient of friction with different sole materials, floors, and contaminants. Appl. Ergon. 35, 499e507. Li, K.W., Chen, C.J., Lin, Ching-Hua, et al., 2006a. Relationship between measured friction coefficients and two tread groove design parameters for footwear pads. Tsinghua Sci. Technol. 11 (6), 712e719.
241
Li, K.W., Wu, H.H., Lin, Y.C., 2006b. The effect of shoe sole tread groove depth on the friction coefficient with different tread groove widths, floors and contaminants. Appl. Ergon. 37, 743e748. Moore, D.F., 1972. The Friction and lubrication of elastomers. In: Rayno, G.V. (Ed.), International Series of Monographs on Material Science and Technology. Pergamon Press, Oxford. Powers, C.M., Blanchette, M.G., Brault, J.R., et al., 2010. Validation of walkway tribometers: establishing a reference standard. J. Forensic Sci. 55 (2), 366e370. Tisserand, M., 1985. Progress in the prevention of falls caused by slipping. Ergonomics 28 (7), 1027e1042. Wilson, M., 1996. Slip resistance characteristics of footwear solings assessed using the SATRA friction tester. J. Test. Eval. 24 (6), 377e385. Wilson, M., 1990. Development of SATRA slip test and tread pattern design guidelines. In: Gray, B.E. (Ed.), Slips, Stumbles, and Falls: Pedestrian Footwear and Surfaces. American Society of Testing and Materials, Philadelphia, pp. 113e123.