Tyre Wear under Urban, Rural, and Motorway Driving Conditions at Two Locations in Spain and China

Abstract

The recently introduced Euro 7 emissions standard regulation foresees the addition of abrasion limits for tyres sold in the European Union. The measurement procedures for tyre abrasion are described in the newly introduced Annex 10 of the United Nations (UN) Regulation 117. However, the limits are not yet defined as there is no data available regarding the new procedure. For this reason, a market assessment campaign is ongoing under the auspices of the UN Task Force on Tyre Abrasion (TFTA). Recent reviews on the topic also concluded that there is a lack of studies measuring the abrasion rates of tyres. In this study, we measured the abrasion rate of one tyre model at two different locations (Spain and China) with the aim of deep diving into possible influencing factors. Additionally, wear rates were studied separately for urban, rural, and motorway routes to get more insight into the impact of the route characteristics. The abrasion rates varied from 22 mg/km to 123 mg/km per vehicle, depending on the route (urban, rural, motorway) and ambient temperature. The overall average trip abrasion rates were 75 mg/km and 45 mg/km per vehicle at the two locations, respectively. However, when corrected for the different ambient temperatures, the rates were 63 mg/km and 60 mg/km per vehicle, respectively. The impacts of other parameters, such as driving dynamics and road surface, on the final results are also discussed. The average tread depth reduction was estimated to be 0.8–1.4 mm every 10,000 km.

1. Introduction

The wear of a tyre is an important topic due to its impact on the service life of the tyre and its impact on the environment. Abrasion rates of tyres have been reported in the literature. Studies before the year 2000 were summarised in a review study in 2004 [1]. Studies from 2000 to 2024 were summarised in a review study in 2024 [2] or other reviews on the topic [3,4]. The majority of the data points were derived from experimental activities carried out by the General German Automobile Club (Allgemeiner Deutscher Automobil-Club, ADAC), mainly on rural and motorway roads [5]. The average abrasion rate at vehicle level was calculated at 118 mg/km (all studies) or 100 mg/km (EU studies) with, however, a big variability depending on a wide range of parameters. The large variability is not only due to the different tyres that were measured [6] but also due to other parameters such as vehicle loading, driving style, and environmental conditions [7,8]. Studies have shown the impact of ambient temperature (e.g., [9]). Vehicle loading is well known and studied, showing, in most cases, an almost linear relationship with the abrasion rate for a specific tyre [10,11]. The driving dynamics can have a significant impact, typically examined in the function of longitudinal and lateral accelerations [12]. Other indices have been used, such as driving severity number (DSN) [13] and relative positive acceleration (RPA) in the real-driving exhaust emissions regulation United Nations Regulation 168 [14,15,16]. Speed and acceleration are typically determined by a global navigation satellite system (GNSS) or accelerometer [17]. The road surface can also result in different abrasion rates [18,19]. For the above reasons, urban and motorway driving have quite often big differences when it comes to tyre abrasion. Urban driving is characterised by stop-and-go and accelerations, thus resulting in higher tyre abrasion. However, according to our knowledge, no studies have examined the abrasion separately under urban, rural, or motorway driving. In an older study [1], tyres driven on motorway roads had lower wear than tyres driven on rural or suburban roads. Furthermore, comparisons of the same tyres at different locations are lacking.

Euro 7 standards regulation foresees the introduction of a limit for the abrasion rate of tyres [20]. The procedure is described in Annex 10 of United Nations Regulation 117 [21] (from now on UNR 117). A similar procedure is described in the draft ISO/DIS 18511-1 “Tyre abrasion rate measurement methods” [22]. The method practically is an on-road convoy method with two (minimum) to four (maximum) vehicles. One of them is fitted with reference tyres. The vehicles drive 8000 km under normal driving and environmental conditions (i.e., they fulfil the boundaries described in the regulation). The route typically consists of motorway and rural roads. The driving dynamics are checked with the standard deviation of the accelerations. The abrasion limits have not yet been defined, and there is an ongoing market assessment to determine the proper definition of the current situation and possible future targets. Furthermore, the European Commission intends to introduce a metric indicating the service life of a tyre in the context of the tyre labelling regulation. One proposal is to measure the tread depth at the end of the 8000 km and extrapolate it until the minimum tread depth (1.6 mm). However, no tread depth measurement procedure has been prescribed yet [23]. For the service life of a tyre, in the United States of America (USA), a similar approach is used, the Uniform Tire Quality Grade (UTQG) rating [24,25,26,27].

The aim of this study is to provide a detailed characterisation of the urban, rural, and motorway roads’ wear of tyres of the same model at two different locations by applying similar testing protocols. In addition to mass loss and tread depth reduction, information regarding tyre temperature, asphalt temperature, and Shore A hardness (a method to characterise how resistant materials are to localized deformation or indentation) [28] is provided. The impact of various parameters on the results is discussed. Different expressions of driving dynamics are also compared. The key novelty of this study is that testing the same tyre model at two different locations allowed for a better understanding of how various parameters, not related to the tyre, affect its abrasion. Equally important is that it has been demonstrated that urban conditions are linked to higher tyre wear. These elements will add a better understanding of tyre abrasion and the parameters that affect it.

2. Materials and Methods

2.1. Overview

The tests were conducted at the premises of two laboratories: IDIADA (ID) in Spain and LINGLONG (LL) in China. Two conventional (i.e., only with internal combustion engine) sport utility vehicles (SUVs) with similar characteristics (

Table 1. Characteristics of the vehicles (only internal combustion engine).

Parameter	ID	LL
Manufacturer	Ford	Geely
Vehicle name	Escape	Lynk & Co 06
Engine capacity	1.5 L	1.5 L
Driving axle	FWD	FWD
Power	135 kW	133 kW
Wheelbase	2710	2640
Curb weight	1494 kg	1430 kg

FWD = front wheel drive.

) were fitted with the same brand tyres: Linglong M+S 225/60 R18 100V Batman A50 SUV. The tyres at ID had been worn at a proving ground for 3000 km, while the tyres at LL were new. After a run-in period in motorway conditions, driving continued on urban, rural, and motorway roads; each of the four parts was approximately 500 to 1000 km, depending on the location. The abrasion rate of the tyres was determined by weighing the tyres at the beginning, at the end, and in between. Details are given in the following paragraphs.

2.2. Locations

The LINGLONG (LL) on-road tests took place at the end of March and beginning of May 2024 around the city of Zhaoyuan in China. The IDIADA (ID) tests took place in June 2024 around the city of Santa Oliva in Spain.

2.3. Test Protocol

The procedure, in general, was the following:

• Vehicle alignment and preparation (loading)
• Tread depth measurements (with air in tyres)
• Tyre assembly weighting (without air in tyres)
• Run-in driving
• Repeat for urban, rural and motorway driving

2.4. Vehicles and Loading

Table 1 summarises the characteristics of the two SUV vehicles, which were similar in terms of dimensions, weight, and shape.

Table 2. Tyre loading conditions. Numbers in brackets refer to percentages of the tyre load index (100), which corresponds to 800 kg.

Parameter	ID	LL
Front Left (FL) tyre load	490 kg (61%)	496 kg (62%)
Front Right (FR) tyre load	460 kg (58%)	457 kg (57%)
Rear Left (RL) tyre load	361 kg (45%)	364 kg (46%)
Rear Right (RR) tyre load	333 kg (42%)	337 kg (42%)
Total vehicle load	1644 kg (51%)	1654 kg (52%)
Front/rear weight distribution	58%/42%	58%/42%

summarises the loading conditions. While the load distribution is representative of FWD vehicles (58%/42% vs. 56%/44%) [21], the overall load is lower than the recommended in UNR 117 for such testing (51% vs. 67%). Therefore, current tests cannot be characterised as UNR117 compliant.

The vehicle alignment was checked with the cars unloaded (Hunter Engineering, Bridgeton, MO, USA) and was within the vehicle manufacturers’ specifications.

2.5. Tyres Measurement Procedures

The car was parked in a garage for a few hours to avoid rain or direct exposure to the sun. The tyre pressure (2.4 bar) was set and/or measured cold. Each laboratory applied the following instructions.

2.5.1. Disassembly and Cleaning

• Record the total wheel balancers installed on each rim.
• Disassemble each wheel from the vehicle.
• Remove stones or any other dirt from the tyre with appropriate tools.
• Clean the tyre with pressurised air.
• Clean the rim surfaces with a dry cloth to remove dust or other small particles.

2.5.2. Tread Depth Measurement

• Measure tread depth at each main groove at four pre-marked positions every 90°. For four main grooves, 16 measurements were taken for each tyre. Each measurement was repeated at least twice, and the average value was used to calculate the tread depth reduction.

2.5.3. Shore A Hardness Measurement

• Measure Shore A hardness at the surface of the tyre just outside the main grooves and in the middle at the four pre-marked positions every 90° (12 measurements were taken for each tyre). Each measurement was repeated at least twice, and the average value was used for calculations.

2.5.4. Wheel and Tyre Assembly Weight Measurement

• Set tyre pressure to 0 bar (remove valve core completely).
• Verify no balances are lost from rims.
• Repeat weight measurement four times per wheel, rotating the wheel 90° each time.
• The wheel is weighted in the same conditions (with hub cover and with valve cab assembled) every time.

2.5.5. Reassembly

• Re-set tyre testing pressure (2.4 bar).
• Re-install wheels to the vehicle in the same position. Do not rotate position; for example, the front-left tyre must always be in the front-left car position.

2.6. On-Road Measurement Procedures

The accelerations were measured with accelerometers installed inside in the middle of the vehicle (ID: PCAN GPS IPEH-002110 from PEAK-System Technik GmbH, Darmstadt, Germany) or high precision GNSS (LL: Drift-Box from Racelogic, Buckinghamshire, UK). The tyre and asphalt temperatures were measured with sensors (ID: IRTS-120-V2 and IRTS-120-V3 from IZZE Racing, Scaggsville, MD, USA; LL: Fluke, Washington, DC, USA) before, during, and after the trips. The ambient temperature was measured with a thermocouple installed on the top and right side of the vehicle.

2.7. Instruments Accuracy

Tyre inflation pressure was measured with a ±5 kPa accuracy manometer. The complete wheel and tyre assembly was weighted with a ±2 g accuracy scale. The tyre groove depth was measured with a ±0.1 mm accuracy depth gauge. The tyre Shore A hardness was measured using a ±5 accuracy Shore A hardness durometer. The temperature devices were accurate within ±1 °C. The accuracy of the accelerometers was ±0.005 g.

3. Results

The results of the tests are given in the following tables separately for urban (

Table 3. Urban trip characteristics (ID for IDIADA and LL for Linglong).

Parameter	ID	LL
Distance (km)	506	1035
Average speed (km/h)	32.8	41.7
Ambient temperature (°C)	24.4 [20.9–31.7]	14.0 [10.0–18.0]
Asphalt temperature (°C)	34.8 [20.9–55.3]	17.5 [13.0–22.0]
Long. acc. std. dev. (m/s2)	0.70	0.92
Lat. acc. std. dev. (m/s2)	0.72	0.72

Note: (…) for units, […] for range.

and

Table 4. Urban tyre results (ID for IDIADA and LL for Linglong).

Parameter		ID	LL
Tyre temperature (°C)	FR	37.7 [21.2–47.9]	21.0 [13.0–30.0]
	FL	38.3 [22.0–48.0]	21.0 [13.0–31.0]
	RR	34.7 [19.1–44.2]	20.0 [12.0–29.0]
	RL	33.7 [19.7–41.8]	20.5 [13.0–29.0]
Shore A hardness (-)	FR	57.9 [53.7–59.7]	64.6 [62.2–66.6]
	FL	56.9 [53.9–60.3]	64.9 [61.3–67.4]
	RR	57.2 [53.9–59.4]	64.8 [61.7–67.3]
	RL	56.5 [52.6–60.9]	65.7 [63.3–67.9]
Tread depth reduction (mm)	FR	0.05 [−0.13–0.23]	0.29 [0.19–0.41]
	FL	0.06 [−0.05–0.16]	0.24 [0.19–0.30]
	RR	0.13 [−0.04− 0.29]	0.18 [0.05–0.30]
	RL	0.09 [−0.13–0.22]	0.25 [0.13–0.44]
Mass loss (g)	FR	21.6	29.0
	FL	15.6	26.0
	RR	14.4	5.0
	RL	10.8	6.0

Note: (…) for units, […] for range. FL = front left; FR = front right; RL = rear left; RR = rear right.

), rural (

Table 5. Rural driving characteristics (ID for IDIADA and LL for Linglong).

Parameter	ID	LL
Distance (km)	1081	1045
Average speed (km/h)	65.4	53.8
Ambient temperature (°C)	22.1 [15.5–31.0]	13.0 [6.0–18.0]
Asphalt temperature (°C)	32.3 [16.9–53.0]	15.3 [8.0–21.0]
Long. acc. std. dev. (m/s2)	0.65	0.74
Lat. acc. std. dev. (m/s2)	1.02	0.79

Note: (…) for units, […] for range.

and

Table 6. Rural tyre results (ID for IDIADA and LL for Linglong).

Parameter		ID	LL
Tyre temperature (°C)	FR	31.3 [20.1–42.6]	21.0 [6.0–32.0]
	FL	32.1 [22.0–41.2]	22.5 [7.0–35.0]
	RR	28.0 [17.6–37.7]	20.0 [7.0–32.0]
	RL	26.7 [15.4–36.6]	19.5 [6.0–31.0]
Shore A hardness (-)	FR	58.3 [53.8–61.0]	66.0 [63.9–67.3]
	FL	57.7 [53.6–61.0]	65.7 [64.5–67.2]
	RR	57.9 [53.6–60.3]	65.5 [63.8–67.1]
	RL	57.2 [52.9–63.9]	66.1 [63.4–67.4]
Tread depth reduction (mm)	FR	0.10 [−0.06–0.26]	0.17 [0.11–0.30]
	FL	0.06 [−0.13–0.14]	0.17 [0.10–0.25]
	RR	0.05 [−0.06–0.34]	0.17 [0.08–0.29]
	RL	0.02 [−0.13–0.12]	0.12 [0.07–0.21]
Mass loss (g)	FR	31.2	10.0
	FL	22.8	10.0
	RR	19.2	10.0
	RL	13.2	9.0

Note: (…) for units, […] for range. FL = front left; FR = front right; RL = rear left; RR = rear right.

), and motorway (

Table 7. Motorway driving characteristics (ID for IDIADA and LL for Linglong).

Parameter	ID	LL
Distance (km)	1014	504
Average speed (km/h)	105.4	98.0
Ambient temperature (°C)	26.6 [19.0–35.9]	16.0 [9.0–22.0]
Asphalt temperature (°C)	37.7 [12.4–52.3]	22.0 [14.0–30.0]
Long. acc. std. dev. (m/s2)	0.41	0.56
Lat. acc. std. dev. (m/s2)	0.54	0.61

Note: (…) for units, […] for range.

and

Table 8. Motorway tyre results (ID for IDIADA and LL for Linglong).

Parameter		ID	LL
Tyre temperature (°C)	FR	39.3 [20.9–52.0]	22.0 [10.0–34.0]
	FL	40.0 [21.8–51.0]	24.0 [10.0–35.0]
	RR	36.6 [19.1–48.4]	22.0 [12.0–32.0]
	RL	34.1 [19.2–44.9]	23.0 [13.0–32.0]
Shore A hardness (-)	FR	58.5 [55.6–59.9]	64.9 [64.1–65.5]
	FL	57.2 [54.6–60.4]	64.3 [61.0–66.1]
	RR	57.2 [55.2–60.1]	64.1 [62.9–66.0]
	RL	56.9 [54.7–60.1]	64.9 [63.7–66.7]
Tread depth reduction (mm)	FR	-0.04 [-0.16–0.13]	0.02 [0.01–0.03]
	FL	0.03 [-0.13–0.16]	0.02 [0.00–0.04]
	RR	0.02 [-0.26–0.16]	0.02 [0.01–0.03]
	RL	0.03 [-0.05–0.15]	0.03 [0.00–0.07]
Mass loss (g)	FR	8.0	6.0
	FL	8.8	3.0
	RR	1.6	4.0
	RL	4.0	4.0

Note: (…) for units, […] for range. FL = front left; FR = front right; RL = rear left; RR = rear right.

) parts. The first table of each set describes the trip characteristics, while the second one gives the results of the tyre measurements. The abrasion level at vehicle level (all four wheels) for the specific routes, excluding run-in, was 65.8 mg/km for ID and 47.2 mg/km for LL.

In general, the ID tests were conducted at an ambient temperature of 10 °C higher than at LL (24 °C vs. 14 °C). The asphalt temperature was 2.5–6.0 °C higher than the ambient temperature at LL but 10–11 °C higher at ID. On average, the asphalt temperature was 17 °C higher at ID than at LL (while the ambient temperature was 10 °C higher). The rear tyre temperatures were close to the asphalt temperatures (in general ±2.5 °C). The front tyres had temperatures around 4 °C higher at ID but 1 °C at LL.

The average speed was, on average, higher at ID than at LL at the rural (65.4 km/h vs. 53.8 km/h) and motorway (105.4 km/h vs. 98.0 km/h) parts but lower at the urban part (32.8 km/h vs. 41.7 km/h). The standard deviations of the accelerations were higher at LL (up to 37%), except for the lateral accelerations at the rural part (-23%).

The Shore A hardness was around 57.5 at ID and 65 at LL without significant variations during the tests or between front and rear tyres. The tread depth was reduced to <0.1 mm/tyre per part at ID. At LL, the reduction was negligible at the motorway part (0.03 mm), around 0.15 mm/tyre at the rural part, and around 0.25 mm/tyre at the urban part. There was no significant difference between front and rear tyres. The mass loss was higher at the front tyres compared to the rear tyres at both ID and LL. The mass loss was higher in the urban part and lower in the motorway part. For a distance of 500–1000 km, the mass loss per tyre was on the order of a few g up to 30 g per tyre.

4. Discussion

4.1. Comparison with UNR 117

Although this study was not designed to follow the recently introduced UN Regulation 117 (UNR 117) on tyre abrasion, to put the results into perspective, a comparison with the provisions of the regulation is attempted in

Table 9. UNR 117 on-road trip requirements. Note that ‘Rural’ is ‘Regional’ and ‘Motorway’ is ‘Highway’ in UNR 117. D = Distance.

Parameter	Requirement	ID	LL
Tyre condition	New	Used	New
Tyre pressure (kPa)	250	240	240
Load (%)	67 ± 7	51	52
Front/Rear distribution (%)	56 ± 7/44 ± 7	58/42	58/42
Circuit min. D (km)	300	20 + 120 + 250 1	40 + 125 + 250 1
Total D (km)	8000	3135	4143
Urban-like D (%)	>25	{33} 2	{33} 2
Urban long. acc. st. dev. (m/s2)	0.45–0.90	0.63	0.81
Urban lat. acc. st. dev. (m/s2)	0.40–1.20	0.65	0.62
Rural-like D (%)	{>25} 3	{33} 2	{33} 2
Rural long. acc. st. dev. (m/s2)	{0.20–0.75} 3	0.59	0.63
Rural lat. acc. st. dev. (m/s2)	{0.70–1.80} 3	0.92	0.71
Motorway-like D (%)	>35	{33} 2	{33} 2
Motorway long. acc. st. dev. (m/s2)	0.10–0.45	0.35	0.44
Motorway lat. acc. st. dev. (m/s2)	0.15–1.00	0.45	0.44
D with speed <60 km/h	>10%	41	29
D with speed 60–90 km/h	>25%	28	43
D with speed ≥90 km/h	>35%	31	28
Max speed (km/h)	140	134	127
Overall long. acc. st. dev. (m/s2)	0.35–0.55	{0.66} 2	{0.65} 2
Overall lat. acc. st. dev. (m/s2)	0.83–1.03	{0.71} 2	{0.61} 2
Max. long. acc. st. dev. (m/s2) (99.98% of D)	±5	2.8	2.6
Max. lat. acc. st. dev. (m/s2) (99.9% of D)	±5	3.8	3.5
Average temperature (°C)	−3 to 35	24	14
Min/Max temperature (°C) for >90% of D	−7 to 40	15 to 36	6 to 22
Driving under snow conditions (% of D)	0	0	0
Driving under wet conditions (% of D)	<20	0	0

¹ the three numbers refer to the distance of different urban, rural, and motorway roads. ² not applicable because each part (urban, rural, motorway) was driven separately, and the results were weighted equally. ³ not a requirement, based on earlier drafts of the regulation.

.

According to UNR 117, the wear of a candidate tyre is determined by fitting a vehicle with new candidate tyres and comparing their mass loss to the mass loss of reference tyres fitted to another similar vehicle following the same circuit. The circuit consists of one or more loops and is repeated until 8000 km are accumulated.

The biggest difference between our study and UNR 117 was that there was no vehicle convoy, and thus, there was no vehicle fitted with reference tyres. A less important difference is that typically, a loop consists of a mix of urban, rural, and motorway roads. In our tests, we tried to have separate loops of only urban, rural, or motorway roads. The main requirements of UNR 117 are summarised in Table 9. The instruments we used fulfilled the accuracy requirements of UNR 117 and are not included in the table. The requirements for vehicle alignment (toe, camber) are also not included. In the regulation, the values refer to loaded vehicles, while in our case, they were checked with unloaded vehicles and were according to the vehicles’ manufacturers. The tyre pressure for the candidate tyres should be 2.5 bar, while in our tests, it was 2.4 bar.

In general, the key requirements of UNR 117 were fulfilled, with one exception. The standard deviations of accelerations of our overall simulated trip (as an average of urban, rural, and motorway wear) were outside the regulatory boundaries: higher for the longitudinal acceleration standard deviation and lower for the lateral one. It should be added that the simulated value was not calculated according to UNR 117; in UNR117, the value is calculated after filtering the GNSS coordinates and not the accelerometer values. Although GNSS data were available, the quality at ID was not sufficient for proper analysis. Thus, our values have some differences compared to the regulatory values due to the use of accelerometers and the lack of filtering. The lack of filtering resulted in approximately 10% higher values in some cases. Note also that the accelerations are distance-based and not time-based; thus, the values of Table 9 are different from those in Table 3, Table 5, and Table 7. The two methods of standard deviations correlate with a slope of around 0.88 and R² > 0.9. Furthermore, we noticed a big difference in the values calculated using the 10 Hz acceleration data or 1 Hz speed data (see Appendix A for details). Thus, the standard deviation values of our study should be interpreted with care.

It is important to highlight that the driving was ‘normal’ and not aggressive (see Appendix A, Figure A1). There was no traffic in the urban region, something that can also be seen by the average speed. Nevertheless, the assessment of different signals for the characterisation of a trip should be further studied with more instruments and vehicles.

4.2. Abrasion Rates

Figure 1 plots the abrasion rates at ID (blue bars) and LL (red bars). The urban, rural, and motorway driving parts are given separately. The ‘Total’ abrasion rate was calculated as the average of urban, rural, and motorway rates, assuming equal distance driving at the three conditions. This is a valid assumption; for example, the type approval cycle of the vehicle exhaust emissions regulation, the worldwide light vehicles test procedure (WLTP) [29] and the real-driving emissions (RDE) on-road test for light-duty vehicles [14] require equal parts of urban, rural, and motorway driving in terms of distance (not time). Note that the run-in abrasion rates were not considered in the calculations.

The abrasion rates of the three parts were 22 mg/km to 123 mg/km per vehicle at ID and 34 mg/km to 64 mg/km per vehicle at LL. The highest values were measured under urban conditions and the lowest under motorways. This result is not surprising given that urban driving is generally linked to more manoeuvres, start and stop situations, braking events, turnings, and other driving conditions that include higher lateral acceleration and deceleration compared to motorway driving. The rates at LL were, in general, lower than at ID. The urban abrasion rate was 5.5 times higher than the motorway abrasion rate at ID but only 1.9 times higher at LL. This could be due to the lower motorway abrasion rate at ID, which could be due to the lower longitudinal and lateral accelerations. The ‘Total’ rates (i.e., average of urban, rural and motorway) were 75 mg/km and 45 mg/km per vehicle at ID and LL, respectively.

The abrasion rates at the two locations (ID and LL) are not completely comparable due to the different ambient temperatures at the two locations. The ambient temperature, which also affects the tyre temperature, has an effect on the wear of the tyres [9]. For the specific tyre, the impact was estimated to be 2.7% per degree Celsius between 9 °C and 29 °C (for details, see Appendix B) [30]. Applying this correction and normalizing the emissions to 20 °C will make the abrasion rates more comparable. The ID abrasion rates, which were determined at around 24 °C, decreased, while the LL results, which were determined at around 14 °C, increased. The normalized rates are much closer to each other (63 mg/km vs. 60 mg/km).

The absolute levels are lower compared to the values reported in the literature (around 100 mg/km) [2]. However, it should be added that the average loading of the tyres was 51% of their load index. UNR 117 defines that testing should be conducted with a loading of 67%. Assuming a linear relation between load and abrasion rate [11], the normalized at 20 °C abrasion rates would be around 30% higher (82.7 mg/km and 77.3 mg/km at ID and LL, respectively); therefore, much closer to the values reported in the literature.

Figure 2 plots the normalized to 20 °C abrasion rates in the function of the standard deviation of the accelerations. We plot the square root of the sum of squares of the accelerations without any factor for the lateral accelerations. This factor is around 0.2 to 0.4 in many cases, taking into account the different wear of the tyre due to lateral forces [13]. Forcing a linear trendline through zero, the slopes of the two datasets (ID and LL) are relatively close (60 vs. 70). While the LL data have some medium correlation (R² = 0.70), the ID data do not have a good correlation (R² = 0.38). The different slopes of the trendlines, as well as the deviation of the data points from the trendlines, might be due to the different road surface characteristics. Assuming that the difference from the trendline gives the impact of the road surface, this could be up to 45 mg/km (based on the ID points). Appendix C gives examples of the road surfaces at the two locations.

Figure 3 plots the contribution of front tyres to total mass loss or tread depth reduction at the two tests (ID and LL). In terms of mass (blue and red bars), the front tyres contributed 50–85% of the total mass loss, with lower ratios for rural and motorway driving. The average contribution of front tyres to total mass loss for the ‘Total’ trip was around 65%. Considering the tread depth, the reduction was, on average, similar at the front and rear tyres (50%, with a range of 35–70%). However, the ratios based on tread depth reduction have high uncertainty due to the small reductions measured, which were within experimental uncertainty. The higher wear of the front tyres has been reported [1,31,32]; however, we did not find separate urban and motorway front-to-rear tyres wear ratios.

4.3. Tread Depth and Tyre Service Life

Figure 4 plots the tread depth reduction per 10,000 km. In urban conditions, the value was 1.6–2.3 mm, but in motorway conditions, it was only 0.3–0.4 mm. On average, 0.8–1.4 mm was expected to be lost every 10,000 km. The results are in good agreement with recent studies that measured tread depth reduction [31,32]. For a maximum tread depth of 6 mm (initial 7.6 mm minus 1.6 mm, which is the minimum required in the regulation), the service life of the tyre would be 43,000 to 75,000 km. The tread depth reduction was always higher at the LL. The LL tyre was new and was preconditioned (run-in) with only 1000 km at motorway roads. The ID tyre had been severely tested with a high wear test (accelerated wear) on a proving ground for more than 3000 km. Then, the run-in driving on motorway roads was another 500 km. Thus, the higher tread depth reduction values at the LL tests might be due to an incomplete run-in distance.

Another interesting parameter is the mass loss per mm of tread. It was around 300 g/mm at the tests at ID and 108 g/mm at LL. There is no clear explanation for this huge difference. The different temperatures at the two locations only slightly change the density. Another reason could be the different history of the two tyres. The more aggressive run-in testing probably changed the density of the material at the surface of the tyre and the contact area with the road. The relatively small values of the tread depth reduction and, consequently, the high uncertainty might have also contributed to the huge difference.

4.4. Uncertainties

The scale measurement uncertainty was 2 g. This value translates to almost 50% uncertainty of the motorway abrasion rate values at ID but <10% of the rural values at LL. The total trip abrasion rates, though, had measurement uncertainty on the order of <5% due to the higher total mass loss. The weight differences between the tyre weighings and the tyre plus wheels assembly weighings were within ±1 g. For our abrasion rates of around 50 mg/km, the scale measurement uncertainty translated to <3 mg/km. The temperature impact was 2.7% per degree Celsius [30]. For our tests, if the temperature impact is not known (i.e., how much the temperature affects the abrasion), the 10° difference at the two locations would result in a ‘bias’ of around 27 mg/km. The driving dynamics had an effect of 4–5 mg/km per 0.1 m/s² change in the standard deviation of the accelerations (Figure 2). The driving at the two locations was quite similar, with differences of 0.1–0.2 units of standard deviation. Thus, the differences between normal driving styles were around 4–10 mg/km. In our study, we could not assess the impact of the road surface. Indirectly, comparing abrasion rates normalized at the same temperature and dynamics, the impact could be up to 45 mg/km (Figure 2).

Table 10. Estimation of uncertainties and impact of different parameters on the abrasion rates for the specific tyre of our study.

Parameter	Uncertainty	Impact
Scale	2 g	3 mg/km
Ambient temperature	2.7%/°C	27 mg/km for 10 °C difference
Dynamics	5 mg/km per 0.1 m/s2	5–10 mg/km
Road surface	Not studied	5–45 mg/km

summarises our estimations of the impact of different parameters on the abrasion rate results.

The uncertainty of the tread depth measurements was high. Although the measurement uncertainty of the tread depth gauge was relatively small (0.1 mm), the measured changes at the different tyre locations had high variability (0.2–0.3 mm). The absolute reduction was small (<0.1 mm) in many cases (all motorways and sometimes rural and urban parts). After approximately 2500 km, the average total tread depth reduction was 0.17 mm and 0.42 mm per tyre at ID and LL, respectively.

Based on the discussion above, the findings of this study strongly support the use of reference tyres for the assessment of a candidate tyre. On the other hand, the relatively good agreement between two different locations supports a narrower abrasion rate range for the reference tyre at 20 °C compared to the one in the regulation (25–75 mg/km/t).

5. Conclusions

In this study, we measured mass loss and tread depth reduction for the same tyre model under urban, rural, and motorway driving at two locations (Spain, around 24 °C, and China, around 14 °C ambient temperature). The tyre loadings were around 51% of their load index. The abrasion rates were 1.9 to 5.4 times higher under urban driving vs. motorway driving. The front tyres contributed, on average, 65% to the total mass loss. The abrasion rates showed some correlation with the driving dynamics, expressed as the standard deviation of the longitudinal and lateral accelerations.

The overall trip abrasion rates with equal distances of urban, rural, and motorway driving were 75 mg/km and 45 mg/km at the two locations but 63 mg/km and 60 mg/km when normalized to the same temperature of 20 °C. Bringing the rates to 67% of the tyre load index, a value defined in UN Regulation 117, would mean 82.7 mg/km and 77.3 mg/km, respectively. The impact of the road surface was not possible to be studied, but the scatter of the abrasion rates from the mean trendline indicated an impact as wide as 5–45 mg/km.

The tread depth reduction was 1.6–2.3 mm per 10,000 km in urban conditions but only 0.3–0.4 mm with motorway driving. On average, 0.8–1.4 mm was found to be lost every 10,000 km. However, these values have high uncertainty because they are based on measurements of approximately 2500 km, with average total tread depth reduction of 0.17 mm and 0.42 mm per tyre at the two locations.

The driving dynamics indices that were assessed showed a relatively good correlation and supported the use of only one of them. In the UN Regulation 117, the distance-based standard deviation of the accelerations is used. However, the determination of the standard deviation based on different methods (e.g., speed signal, GNSS coordinates, or accelerometers) may have differences. In the current UN Regulation 117, only one method is prescribed (GNSS coordinates).

The not well-defined shares of urban, rural and motorway roads and driving styles might lead to some differences in the results of different laboratories. Similarly, it has been shown that ambient temperature and acceleration can have a big impact on tyre wear. The influence of all these parameters can be minimised by using a vehicle fitted with reference tyres. In general, the findings of this study support the provisions of the recently introduced regulation.