Fuel consumption rates of passenger cars in China: Labels versus real-world

Abstract

Recently, China has implemented many policy measures to control the oil demand of on-road vehicles. In 2010, China started to report the fuel consumption rates of light-duty vehicles tested in laboratory and to require new vehicles to show the rates on window labels. In this study, we examined the differences between the test and real-world fuel consumption of Chinese passenger cars by using the data reported by real-world drivers on the internet voluntarily. The sales-weighted average fuel consumption of new cars in China in 2009 was 7.80 L/100 km in laboratory and 9.02 L/100 km in real-world, representing a difference of 15.5%. For the 153 individual car models examined, the real-world fuel consumption rates were −8 to 60% different from the test values. The simulation results of the International Vehicle Emission model show that the real-world driving cycles in 22 selected Chinese cities could result in −8 to 34% of changes in fuel consumption compared to the laboratory driving cycle. Further government effort on fuel consumption estimates adjustment, local driving cycle development, and real-world data accumulation through communication with the public is needed to improve the accuracy of the labeling policy.

Introduction

China's vehicle population has increased explosively during the past three decades. In 2009, China exceeded the US in vehicle sales and became the largest vehicle market in the world. As a result, the increasing oil consumption of road transport becomes a major concern in China. To date, China has implemented a series of measures to control the oil demand of on-road vehicles. In 2004, China issued its first vehicle fuel economy standard (implemented in 2005), which is a significant step towards a sustainable and environmental friendly transportation for China. In 2006, in order to aid the implementation of the standard, the National Development and Reform Commission of China (China NDRC) (2006) released the fuel consumption rates of more than 400 passenger vehicle models, which began the publicity of vehicle fuel consumption rates to the public.

In Jan 2010, the Chinese government made another significant step—implementing the labeling policy, which requires each new light-duty vehicle (3,500 kg or less in weight) to be labeled with the fuel consumption rate that the government estimated for the model type. The estimates are based on measurement results performed in laboratory under the New European Driving Cycle (NEDC). Meantime, the government reports fuel consumption rates of thousands of light-duty vehicle models on an official website (http://gzly.miit.gov.cn:8090/datainfo/miit/babs2.jsp), which is free for the public to access. The Ministry of Industry and Information Technology of China (China MIIT) is responsible for the implementation of the labeling policy. By September 2010, China MITT has reported fuel consumption rates for approximately 10,000 vehicle models of over 100 auto manufacturers, and the government is updating the database on a monthly basis with more vehicle models.

However, numerous evidences have shown that differences exist between laboratory and real-world fuel consumption rates. Schipper and Tax (1994) examined the data of the US, Canada, and four European countries and found that automobile fuel economy tests tended to underestimate fuel use by 15–25%. Zachariadis (2006) concluded on the basis of the International Energy Agency (IEA) estimates for nine EU countries that the difference between on-road and test fuel consumption has increased in the last two decades from 7% higher for on-road in 1980 to 13% higher in 1997. The reasons for the differences are attributed to the fact that tests do not reflect real-world driving and driving is increasingly becoming less fuel-efficient as congestion gets worse. Smaller gap was also reported. Meyer and Wessely (2009) estimated that the divergence between test and real-world fuel efficiency could be 1.9–5.1% on the basis of web data reported by real-world drivers.

The gap between laboratory and real-world fuel consumption is universal, and it could be the same for China, especially China uses the European driving cycle (NEDC) to estimate fuel consumption values. The NEDC cycle is used for Chinese emission standards initially because China follows the European emissions standards, e.g. Euro I, Euro II, Euro III, etc., then its use extends to the fuel economy measurements and policy. Previous studies have pointed out that the European driving cycle cannot reflect the real driving conditions in China (Wang et al., 2008). Recently, the difference between the government estimates and real-world fuel consumption levels has been noticed by the public in China. The Chinese government needs to act in response to the differences.

The US has implemented the labeling policy for over 30 years. According to the policy, all new cars and light-duty trucks sold are required to have a label showing fuel economy estimated by the US Environmental Protection Agency (EPA). The effectiveness of the fuel economy estimates has been debated for decades. During the past 30 years, the US EPA has been dedicated in providing more accurate estimates. In 1984, the US EPA adjusted dynamometer test results downward by 10% for the city cycle and 22% for the highway cycle to more accurately reflect driving styles and conditions (US Environmental Protection Agency, 2004). Recently in 2006, the US EPA changed the methodology for calculating the city and highway fuel economy label estimates for new passenger cars and light trucks to provide fuel economy estimates to consumers that better reflect real-world fuel economy (US Environmental Protection Agency, 2008).

The labeling policy in China is new, but there is an urgent and practical need for China to evaluate and make an appropriate adjustment to the fuel consumption estimates. However, the current understanding on real-world fuel consumption in China is insufficient to support policy makers. For instance, the real-world fuel consumption level is even unknown in China because China has not conducted official tests or investigations on vehicle fuel use. Previous studies have been focused on laboratory fuel consumption in China (Wang et al., 2010, Wagner et al., 2009, Oliver et al., 2009), and little work has been done for real-world fuel consumption.

Nevertheless, Chinese car owners provide great help in data collection. Right after the government initiated the new labeling policy, Chinese drivers started to share and exchange their fuel consumption estimates on the internet in order to compare the actual fuel consumption levels to those provided by the government. The web data are under real operation conditions. One of the most popular webs is the Auto Channel of the Sohu Website (http://db.auto.sohu.com). To date, people have posted over 100 thousand pieces of real-world fuel consumption rate information for more than 500 car brands on this website. Since China lacks large-scale official investigations on vehicle fuel use, these self-organized surveys are extremely valuable to a better understanding of the real-world fuel consumption of Chinese vehicles.

This study aims to examine the differences between test and real-world fuel consumption of light-duty passenger vehicles in China and provides appropriate suggestions to policy makers to improve the fuel consumption estimates and labeling policy. We collected the fuel consumption rates reported by the government (Ministry of Industry and Information Technology of China, 2010) and Chinese car owners (Sohu, 2010). The former represent the laboratory fuel consumption rates, and the latter can be used as a surrogate to approximate the real values due to unavailability of rigorous statistics. The International Vehicle Emission (IVE) model is employed to simulate the fuel consumption rate changes under the standard test procedure and real-world driving conditions. On the basis of our study, we call for additional government efforts to improve the current labeling policy.