Approach on Utilizing Customer Usage Profile for ALT Design

Approach on Utilizing Customer Usage Profile for ALT Design Azianti Ismail Faculty of Mechanical Engineering, Universiti Teknologi MARA, Malaysia Won Jung, Department of Industrial and Management Engineering, Daegu University, South Korea Email: [email protected] Abstract: Manufacturers would learn important information by collecting and analyzing customer usage profile of the current component captured in the appropriate usage and environmental stresses. When customer usage profile has been collected, there is assessment of what data is valid and useful, what assumptions need to be made and finally to design suitable accelerated life testing (ALT) for the component. This case study of agricultural tractor transmission provides the basis for ALT designs for the component based upon customer usage measurements which to be conducted to estimate the current reliability of the product. Thus, understanding on the environmental and operating conditions of the components use environment is crucial. Keywords: customer usage profile, Accelerated Life Test, agricultural tractor 1. Introduction Usage profile should include a variety of climatic conditions, environmental conditions and working conditions such as crop variety, soil type. It also covers customer usage patterns to determine possible failure modes, geographic location, purpose and handling of the tractors to provide exact use conditions. Campean et al. have discussed on a generic approach to life prediction modeling for automotive components, which aims to establish a correlation between the degradation mechanism, customer usage profile and rig life testing [1]. Ion et al. have mentioned that the reliability of the product can be checked whether it is at the right level by using field load data. It gives the opportunity to take immediate actions in case unforeseen reliability problems occurred at customer’s use conditions [2]. Meeker et al. have mentioned that with good characterization of field use conditions, it may be possible to use ALT results to predict the failure distribution in the field [3]. On the other hand, Attardi et al. have presented case study regarding the reliability analysis of some automotive components based on field failure warranty data [4]. From figure 1, customer usage profile consists of the environment, usage pattern and customer feedback. Variation of working condition such as type of soils, weather pattern such as temperature or humidity and geographical location such as slope or flat surface are considered as environmental influenced factors. Customers’ complaints or suggestions relating to the working conditions or failures occurred would be helpful to more understand the causes of failures. Customer Usage profile will help to develop ALTs that generate the same failure modes, driven by the same failure mechanisms, when compared to actual use environment. Simultaneous channels count capability to record loads, strains, temperatures and environment parameters are then analyzed to get the value for designing ALT. It will help to gain better understanding of the usage profiles and to replicate failure modes those are observed in the field. 1 Figu gure 1: Factors involved in Customer Usage Profile 2. Background of the Study Data has been collected duringg field f visits to four areas under four Provinces (Jeollanam-do), ✂✁☎✄ (Gangwon-ddo) and ✆☎✟ ✄ ✆✞✝ ✄ (Gyeonggi-do), ✠☛✡ ✄ (Gyeongbuk-do). The total of 110 random ran farmers has been interviewed and answered the questi stionnaires. Normal operating conditions for customerr usage u profile have been obtained through surveys and field ld visits. Information in the questionnaires which cover vered the main activities for the tractor and also working attachment at equipment used such as loader, plower,, rotary r or trailer. Other information that included is duration ion of use (hours per month), gear utilization (high, med edium or low) and place of use (paddy fields, live stocks, farm arms). By analyzing the questionnaires in which to gett ccustomer usage profile, it can determine possible failuree modes, m geographical location, purpose and handling ing of the tractors. The equivalent load which includes equi quivalent torque and speed that have been applied onn th the transmission for all the agricultural activities must be calculated. ca For this case study, plower operation has bee been selected to show the detail calculation of the equivalentt damage d load. 3. Usage Profile Data 3.1 Working condition based on the th location The manufacturer produces sev everal model of tractor with different sizes and horsepo power. 78 percent of the customers in those areas involved own ow medium sized tractors in which under 70 hp. Out ut of the 78 percent, 36.4 percent customer own 41 to 50 hpp tractors. t Thus, tractors that below 50 hp are focus for this case study. From figure 2, tractors used in paddy fields fie have recorded higher number of agricultural ac activities which mostly involved with plower and rotary acti ctivities. 2 $ &# # ( $ % & ' Fig igure 2: Usage Time Based on Working Condition 3.2 Seasonal usage due to weather er pattern From figure 3, March to May ay is the highest utilization of the tractors in all of the he agricultural activities. This is due to the spring season in i which most of the agricultural activities start du during this time period. November to February have resulted ted in low utilization of the tractors due to winter season on which is cold and dry. During winter, most of the tractors rs are kept either under shelters or just left them outsid tside exposed to the cold weather. Besides that, most of thee tractors have been used as trailers during Augustt to October in which to transport goods after harvesting. Annual An tractor usage time is 160 hours which are mos ostly used during spring and autumn seasons. ! $ % & ' " # ) * ) & " " * + ( , - Figure 3: Usage Time Based on Month 3.3 Utilization of gear based on agricultural ag activities Utilization of gear is different nt based on activities involved. For this case study, plo lower, rotary and loader activities utilize on medium gears. rs. Meanwhile, trailer utilizes mostly all the gears from rom high to low. Plower activity is chosen as the example of this t case study. Figure 4 shows the utilizationn of power consumption based on type of gear. From th the figure 4, gear M2 is the most common gear that is used ed during plower activity. Thus, gear M2 is the main gear ge that will be further investigated. Next step is to determ rmine which axle that needs to be focused either the fr front or rear axle. From figure 5, rear axle has contributed higher hi torque value compared to the front axle. 3 ☞✍✌ ✎ ✏✑✌ ✏ ✏✑✌ ✒ ✏✑✌ ✒ 0& . $ /. .! *1 .! ✓☞ ✓✔ ✕✖✎ ✘✚✙✞✛✢✜ $ ✕✗✏ Figure 4: Power Train Consumption on Plower Activity 1000 900 800 700 600 500 400 300 200 100 0 882.7 848 856.8 173.5 177.3 160.6 809.9 178.2 L3 ✣✥✤✧✦✩★ L4 M1 Rear Axle M2 Front Axle Figure 5: Rear and Front Axle Torque 4. Duty Analysis Based on linear damage accumulation rule (commonly called the Miner’s rule) was proposed by Palmgren in 1924 and later by Miner in 1945, it has widely been used to calculate the equivalent damage loads and fatigue strength of the components [5]. The cumulative damage law, Palmgren-Miner damage calculations has been performed to obtain the equivalent torque and speed applied on the transmission. Duty analysis is to create similar conditions of use which is determined from field load history. Field load history is based on usage conditions and activities involved. Duty analysis is transmission equivalent load determined by analyzing field data in which include the load measurement for agricultural activities such as plower, rotor, loader and trailer. The equivalent load which includes equivalent torque (Pe) and speed (ne) that have been applied on the transmission for all the agricultural activities must be calculated. Using the formula for Pe and ne, the equivalent damage load performed on tractor transmission during agricultural activities such as loader, trailers, rotor and plower could be calculated [6]. 1 λ Σ h i n i Pi λ Pe ( ne = ( Pe = Σ f i Pi 4 λ ) (1) 1 ) λ (2) Table 1 shows the channels count capability to record loads, strains, temperatures and environment parameters. Pressure of hydraulic and measurements of temperature are used to monitor for any abnormalities during the torque and speed data collection in the laboratory. From figure 6, range of rear axle torque, Pi is fixed between 0 to 1400 kg.m. This range has been collected through data recorded during field visits and surveys. Speed, ni and ratio of time for equivalent rate, hi are determined from the range of rear axle torque and also through laboratory measurement. Table 1: Channels of recordings data User Label PTO Shaft Rotation Right Rear Axle Rotation Engine Speed Left Rear Axle Rotation Hydraulic Oil Pressure Engine oil temperature Transmission fluid temp Exhaust gas temp Engine coolant temp Air cleaner intake temp Ambient temp Physical Channel CH01 CH02 CH03 CH04 CH05 CH06 CH07 CH08 CH09 CH10 CH11 Range 0 to 18,000 0 to 9,000 0 to 35,300 0 to 9,000 0 to 250 -270 to 1,372 -270 to 1,372 -270 to 1,372 -270 to 400 -270 to 1,372 -270 to 400 Units rpm rpm rpm rpm bar deg Celsius deg Celsius deg Celsius deg Celsius deg Celsius deg Celsius The maximum rotational speed of the rear axle mounted under the chassis dynamometer driven by the constant load torque by each gear is added to the test. Thus, by using data from the figure 6 together with equation (1) and (2), equivalent damage load for torque and speed can be calculated for the transmission during plower activity. Table 2 shows detail calculation of the equivalent torque and speed for the transmission based on load data. ✯✷✯✷✯ ✵ ✯✷✭ ✪ ✸ ✪✬✫✮✭ ✯✱✰ ✹✷✯✷✯✷✯ ✯✷✭ ✪ ✴ ✯✷✯✷✯ ✯✷✭ ✶ ✸ ✳✷✯✷✯✷✯ ❉ ✸ ✯✷✯✷✯ ❇❈ ✲ ✯✷✯✷✯ ✯✷✭ ✶ ✪✱✯✷✯✷✯ ✴ ✭ ✪✮✰✚✵ ✶✷✯✷✯✷✯ ✲ ✭ ✳✱✰ ✭ ✶✱✰ ✯✷✭ ✫ ✸ ✫✮✯✷✭ ✳✱✰ ✯✷✭ ✫ ✸ ✭ ✪✮✰ ✲ ✭ ✫✑✰ ✫✮✯✷✯✷✯ ✶✷✭ ✹✱✰ ✯✷✭ ✯ ✸ ✯✷✭ ✪✮✰ ✯ ✯ ✻✽✻✽ ✼ ✾✻❀✻❀ ✿ ✻✽ ❁ ❂✻❄✻✽ ❃ ❅✻✽❆✻ ✺✻ ✻ ✻ ✻ ✻ ✻ ✻ ✻ ✻ ✻ ✻ ✺✻ ✺ ✺✻ ✻ ✺ ✻ ✻✼ ✺ ✻ ✻✾ ✻✿ ✻ ✺ ❊●❋✬❍✬■❑❏▼▲●◆❖❊●❏✷❋✬P❘◗❘❙❯❚ ❏✚❱●▲❲P ❳❲❨✬❏ Figure 6: Rear Axle Torque Duty Cycle for Plower Activity (M2) From calculation in table 2, equivalent torque Pe is calculated to be 778.3 kg.m and equivalent speed, ne is 9.7 rpm. This information can be used to calculate acceleration factor to design accelerated life tests for the transmission. 5 Table 2: Calculation Equivalent Torque and Speed for Plower Activity Pi Torq (kg.m) 100 200 300 400 500 600 700 800 900 1000 1100 1200 1300 1400 ni Speed (rpm) 5.77 5.99 5.87 12.52 12.13 12.50 12.38 12.33 12.28 12.28 12.12 12.21 12.14 11.41 hi time (%) 31.0 4.6 4.7 4.1 4.3 7.3 9.2 10.7 10.6 5.3 4.1 2.8 1.0 0.3 nihi Speed (rpm) 1.79 0.28 0.28 0.51 0.53 0.91 1.14 1.31 1.30 0.65 0.50 0.35 0.12 0.03 fi (%) fi P i 3 nihiPi3 18.5 2.9 2.8 5.3 5.4 9.4 11.7 13.6 13.5 6.7 5.1 3.6 1.2 0.3 1.8E+05 2.2E+05 7.6E+05 3.3E+06 6.8E+06 2.0E+07 4.0E+07 6.9E+07 9.8E+07 6.7E+07 6.8E+07 6.1E+07 2.6E+07 8.7E+06 1.7E+05 2.2E+05 7.4E+05 3.2E+06 6.5E+06 1.9E+07 3.9E+07 6.7E+07 9.5E+07 6.5E+07 6.6E+07 6.0E+07 2.6E+07 8.4E+06 Equiv Torq (kg.m) Equiv Speed (rpm) 778.3 9.7 5. Conclusion Measurement and analysis of customer usage profile is used to determine the equivalent damage load that applied to the life of the component. It is very important to determine the accurate equivalent loads that portray the real use conditions of the component in the field. It also provides the basis for accelerated life test designs for the component based upon customer usage profile from usage. Normal operating condition values have been obtained for closely demonstrating the component life. It will help researcher to design ALT more accurate and results obtained from the tests will be useful in predicting the reliability of the component. References [1] Campean, I.F., Day, A.J., and Wright, S. (2001). Camshaft timing belt reliability modeling, Reliability and Maintainability Symposium, 2001. Proceedings. Annual, vol., no., pp.377-383, 2001 [2] Ion R.A., Petkova, V.T., Bas H.J. and Sander P.C. (2007). Field reliability in prediction in consumer electronics using warranty data, Quality and Reliability Engineering International, vol. 23, pp. 401-414. [3] Meeker, W. Q., Escobar, L. A. and Hong, Y. (2009). Using accelerated life tests results to predict product Field reliability. Technometrics, 51, pp. 146-161. [4] Attardi, L., Guida, M. and Pulcini, A.G. (2005). Mixed-Weibull regression model for the analysis of automotive warranty data, Reliability Engineering & System Safety, Volume 87, Issue 2, February 2005, pp. 265-273. [5] Zaretsky, E. V. (1997). A. Palmgren Revisited: a basis for bearing-life prediction. Kansas City, MO: NASA Technical Memorandum 107440, May 18-22, 1997. [6] Ismail, A. and Jung W. (2011). Estimation of Equivalent Damage Load for ALT Design, KSIE Conference Spring, 28 May 2011, Gumi, South Korea, pp.114-119. 6