详情

原文传递 Transportation Leapfrog: Using Smart Phones To Collect Driving Data and Model Fuel Economy in India

题名：	Transportation Leapfrog: Using Smart Phones To Collect Driving Data and Model Fuel Economy in India
作者：	Gopal, A.; Schewel, L.; Saxena, S.; Phadke, A.
关键词：	ENERGY CONSERVATION##CONSUMPTION##Smart phones##Driving data##Fuel economy##India##ENERGY PLANNING##POLICY##ENERGY STORAGE##Utilization analysis##Economy analysis##Transport infrastructure##Car ownership##Telecommunications##Mobility analysis##Vehicle monitoring systems##Urban Transportation Planning Package##Pune city(India)##Algorithms flow##Transportation Leapfrog##
摘要：	Car ownership in India is expected to skyrocket in the next two decades (1). India is projected to become the world’s third largest auto market after the US and China by 2030 and possibly overtake the US by 2035 (2). Most importantly, this demand is primarily due to rising incomes and cannot be easily averted through aggressive Avoid-Shift (A-S) policies because car ownership is dictated by more than a simple desire for convenient mobility (3). Automakers recognize the huge emerging market both in India and China and are gearing up to supply them. However, if the car ownership projections come true, India alone will be responsible for almost 8% of global transportation greenhouse gas emissions by 2030 (4, 5) and will need to import more than 85% of its crude oil (1). In addition, India already has the highest annual road accident deaths in the world (6), some of the world’s worst air pollution from transport, and severely underdeveloped transport infrastructure (7). Thus, the social and environmental externalities from this car boom need to be aggressively and cost-effectively mitigated starting immediately. To design effective measures policymakers, academics, urban planners and civil society need excellent data from Indian transportation. Unfortunately, there is very little macro data on the Indian transport sector (8) and virtually no useful data on mobility behavior and demand (4, 9). The traditional approach to transport data collection follows a hardware intensive approach with installation of on-road sensors, laser and vehicle monitors, specialized in-vehicle loggers, etc. Developed nations such as the U.S. have invested tens of billions of dollars in such data collection infrastructure for transportation (10). Current hardware approaches are very expensive. In the US, each traffic monitoring device to be used on a single intersection costs between $2,000 (for a simple loop traffic counter) and $24,000 (for machine vision), plus installation costs and $2,000-$4,000/year for maintenance (11). These costs do not include the installation and maintenance of a data management system. India had approximately two million kilometers (km) of paved roads in 2008, according to the World Bank (12). If just 20% of these kilometers were monitored for simple vehicle speeds and counts, the hardware costs would rise to $4 Billion (assuming an average of $ 10,000/device and one device per km). India does not have the time, or the capital resources, to follow such a hard path that collects only mdimentary information. Fortunately, the extremely rapid development of India’s mobile telecommunications infrastructure provides us with the opportunity to get even better transportation data than traditional approaches at much lower costs. Several states within the U.S. have found that the costs of using vehicle probes (dedicated vehicles, usually commercial, with installed speed monitoring equipment) are about one-fifth to one-fourth that of dedicated hardware. In this paper, we describe an innovative transport data collection framework that is cheaper and able to gather more data than the probe vehicle approach. Our approach piggybacks on the great Indian telecommunications leapfrog (13, 14), to catalyze an equally significant leapfrog in transportation data acquisition and analysis. Specifically, we describe the technical and economic details of a pilot project in which we use commercially available smart phone apps to collect per second data on speed, acceleration, GPS location and inclines for cars in the city of Pune that is instantly uploaded by 3G and then prepared for analysis using advanced noise filtering algorithms. The data we collect has numerous applications that range from systems engineering design of automobiles to urban transportation planning and management. In this paper, we choose to highlight an application that can substantially improve the labeling test procedure for India’s proposed passenger car fuel economy standards (15).
总页数：	15
报告类型：	科技报告