Efficiency Models in Roads
The importance of determining road efficiency is an imperative area that has been examined by various researchers. Dollery & Wallis (2001) noted that an increasing number of studies have employed the nonparametric technique of data envelopment analysis (DEA) to investigate road sector efficiency. Odeck (2008) studied that DEA is a typical benchmarking analysis commonly used in econometrics to estimate the efficiency of road transport. The aim of DEA is to calculate an efficiency frontier measured as relative performance of different units called DMU (Decision Making Units) in terms of distance per unit to the ideal frontier, constructed using observed input and output data (Brebbia, 2013).
Traditionally, the evaluation of the road efficiency was mainly based on the indexes of V/C ratio and ratio speed, aiming at improving traffic performance. Mao (2010), however, noted that the improvement can guarantee neither the decrease of the spatiotemporal cost of single vehicle, nor the increase of the transportation efficiency. Based on efficiency analysis theory, using DEA analysis, researchers can build a CCR model to evaluate an alternative scheme and to analyze and diagnose the ineffective scheme (Mao, 2010). Bogetoft & Otto (2011) and Weiermair & Perlman (1990) asserted that from the perspective of the road efficiency, the road as a business object also meets general characteristics of production systems where DEA model is widely used. With the increase of traffic resources input, Mao (2010) established that the level of traffic operation will increase, the operating cost of an individual vehicle will decline, and then the formation of economics will be produced.
The objective of road efficiency models are based on meeting the largest possible number of traffic demand with less open road space and shorter traffic delays (Ozbek, de la Garza & Triantis, 2010). This implies that considering improving operational efficiency of the road by reducing the resource consumption level of individual vehicles, the funds available, time and space resources of the road, environmental pollution price and other resources are the most beneficial improvements that can be achieved through DEA model (Mao, 2010).
According to Odeck (2008) and Ramanathan (2004) the objective of DEA is to compare performances of different urban networks in order to provide technical support to policy makers in the choice of actions to be implemented to make the systems efficient urban roads. Brebbia (2013) says that DEA analysis allows highway agencies to calculate for each road network a value of relative efficiency on the basis of which the networks are ranked distinguishing efficient from inefficient ones. Brebbia (2013) further noted that the efficiency of each network is only meaningful within the context examined, in relation to the model chosen and to the sample units considered. The one only needs to introduce a new network or to change the model characteristics from input to output oriented.
Dollery & Wallis (2001) in their research used DEA to measure the relative efficiency of Ontario’s highway maintenance patrols. The inputs in this case were patrol service and capital expenditures and the outputs were stipulated in terms of the characteristics of the roads serviced and an accident prevention factor. In an effort to address road efficiency Cooper, Seiford & Zhu (2011) utilized a comprehensive framework that measured the overall efficiency of road and highway operations. They used DEA model, designed to consider the effects of environmental and operational factors on such overall efficiency. They established that DEA can be used for the relative efficiency evaluation of multiple homogeneous decision making units.
In addition, Cooper, Seiford & Zhu (2011) stress the fact that it has become imperative for road sector organizations to rationalize the operating costs and to improve the quality of the services offered. They obtain measures of pure technical, scale and overall efficiency of both public and private agencies. Zhang (2010) compared road efficiencies operating in the USA, whereas Boile (2001) evaluated road transport efficiency using the DEA model. According to Zhang (2010) and Boile (2001), the empirical analysis of DEA model contains an index of road surface defects and a measure of roughness for both urban and rural roads. They therefore noted that DEA is an effective method measuring road efficiency based on decision making units of the same type and multi-input and multi-output for relative efficiency and benefits.
Consequently, Bhagavath (2009) studied that the achieved efficiency degree, using DEA by each road network, is meaningful only in the context in which it has been measured and then only in relation to the specific model and sample units considered. Brebbia (2013) also examined that simply introducing a new road network or changing the model characteristics from input to output oriented or inputs will produce different efficient networks or different efficient values. Brebbia (2013) noted that the DEA technique is well suited for analyzing and comparing road network efficiencies. The DEA model should be chosen on the basis of the researcher’s ability to retrieve the data necessary for processing highway efficiency (Charnes, 1994).
In the evaluation process of the DEA model, Bogetoft & Otto (2011) asserted that it is difficult to distinguish the input variables (cost variables) and the output variables (efficiency variables). In this context, Mao (2010) noted that the analysis of the DEA’s main influence factor, the decision making units, traffic volume control and control lane are the main elements in the DEA model use. He noted that DEA is used to road efficiency, make up for the shortcomings of traditional assessment methods and analyze intrinsic link among elements (Mao, 2010). Palma (2011) also noted that the productivity index from DEA should be regressed against various spatial factors and should be suggested that the location of the highway plays an important role in determining its efficiency.
Empirical analysis of road sector efficiency using DEA suggests that it is a unique product of complex non-discretionary inputs and outputs and constraints, multiple inputs and outputs factors. Since the DEA model shows the cases of inefficient road networks, Brebbia (2013) indicated that the input values, that should be adopted in order to achieve efficiency, should be minimized whereas the output values should be maximized. Cook & Zhu (2005) noted that the advantage of using the DEA framework is capable of handling noneconomic factors like number of accidents, maintenance dollars per day, cars per day, average age of pavement and allows for measurement of such factors on different scales. Bhagavath (2009) argued that the DEA model is particularly suited for determining the efficiency of roads since factors such as traffic intensity and safety parameters are important part of road transport.
In conclusion, efficiency measurement and benchmarking in road transport is an important topic whether one is interested in comparing efficiency of different road networks or learning to improve efficiency performance of highways. One complication here is that agencies, involved in measuring the efficiency of road sector, provide multiple outputs sharing a common set of inputs in producing these multiple outputs. Calculation of relative efficiency scores with the DEA model generates insight into the performance of highways on various dimensions thus guiding the required action. The benefit of using the DEA model to evaluate road transport efficiency is that it only requires physical data, and not financial or nominal data, free of a priori assumptions on functional forms and applicable to multi-output highways. The weakness of the DEA model is that is it is extremely sensitive to outliers and generates multiple best performers and is in efficient, because it uses only a subset of observations in identifying production possibility set. DEA can be combined with other approaches to separate and measure the technological advances that can be used in highway efficiency improvement over time.