Policy Improvements for Winter Road Maintenance in South-East Europe

Abstract

Weather, as an important factor for traffic conditions, will have increased significance in South-East Europe (SEE) due to ongoing climate change. Consequently, SEE countries in transition will need to reconsider their winter road maintenance (WRM) practices. This article aims at providing a starting point for improving WRM practice. Focusing on Serbia, the article first provides a critical evaluation of existing WRM state of the practice. The core issue identified is a lacking practice in economic evaluation of WRM investment decisions. In addition, the article presents methodology and an example case study for economic evaluation of WRM investment decisions based on cost–benefit analysis. Results indicate an economic justification for investing in improved WRM practice. Finally, the article concludes that WRM policy improvements need to base on knowledge transfer from international best practice, improvement in performance measurement, methods, and technologies, accompanied with an essential change in institutional culture to accommodate policy learning.

Keywords

cost–benefit analysis winter road maintenance economic evaluation Serbia economies in transition policy learning climate change

Introduction

Weather is one of the most significant factors affecting road and traffic conditions. First, heavy precipitation and snowfall can result in significant speed and capacity reduction (Cao, Thakali, Fu, & Donaher, 2013; Enberg & Mannan, 1998; Kwon, Fu, & Jiang, 2013; Maze, Agarwai, & Burchett, 2006; Sabir, van Ommeren, Koetse, & Rietveld, 2008) and increase in the standard deviation of speed (Alppivuori, Leppänen, Anila, & Mäkelä, 1995). Consequently, users experience non-recurrent delays and increased travel time (TT; Tsapakis, Cheng, & Bolbol, 2013; Williamson & Estis, 2007). Second, weather can have a significant impact on traveler’s safety, by increasing traffic accident rates or simply by increasing the risk of an accident (Andersson & Chapman, 2011; Andreescu & Frost, 1998; Eisenberg, 2004; Eisenberg & Warner, 2005; Maze, Crum, & Burchett, 2005). On one hand, the influence on these parameters originates from reduced road friction. On the other hand, an even greater effect on flow and safety might be originating from the effect on driver’s visual tracking of the road delineation and surrounding vehicles (Keay & Simmonds, 2006; Koetse & Rietveld, 2009; Kwon et al., 2013; Usman, Fu, & Miranda-Moreno, 2012; Wallman, Wretling, & Öberg, 1997).

Considering the effects on efficiency and safety, countries in the cold regions regularly prepare and perform winter road maintenance (WRM). WRM includes a complex set of activities for road maintenance in icy and snowy conditions, which frequently vary from one country to the other. For example, WRM activities include mechanical and manual snow removal, deicing using mixture of road salt and crushed stone aggregate, the use of anti-icing chemicals, provision of traveler information, and so on. Most prominent practices for WRM originate from northern United States, Canada, Sweden, Finland, Norway, the United Kingdom, Netherlands, Germany, and Japan. A common misconception is that besides countries in the colder regions, rarely do other countries need or perform WRM. One of those regions, commonly considered not to need WRM due to warmer climate, is South-East Europe (SEE). Contrary to this common perception, SEE countries do perform winter maintenance, as we will later exemplify. However, there is one significant change on the horizon which will substantially affect WRM policies in SEE countries—climate change.

Current climate of this region is a combination of temperate continental and Mediterranean subtropical climate (Gocic & Trajkovic, 2013). However, climate projections for SEE show that this region is particularly sensitive to climate variability (Berkhout et al., 2009; Bernstein et al., 2007). SEE belongs to a region that is highly vulnerable to climate change (“Belgrade Initiative,” 2007). Consequently, the effects of snow and low temperatures can also be experienced in areas that typically had relatively mild and snow-free winters. In addition, despite the predictions of smaller average winter precipitation, there might be individual precipitation periods with larger amounts of precipitation (“South East European Climate Change Framework,” 2008). Observed tendencies support these predictions, as there has been an upward trend in extreme winter temperatures (Unkašević & Tošić, 2008) and a rise in the number of cold days in the fall (Unkašević & Tošić, 2013).

The threat and evidence of climate change has additional implications for SEE region (Figure 1). These implications originate from the fact that most of the countries in SEE region are countries in transition (e.g., Bosnia and Herzegovina, Former Yugoslav Republic of Macedonia, Montenegro, Serbia, etc.). This fact has twofold consequences. First, SEE countries have greater resource constraints for WRM activities. Lessons learned from agencies operating in the colder regions countries tell us that WRM activities can be resource intensive. For example, in the United States, WRM operations are estimated to account for US$2 billion each year (Minsk, 1998), consuming as much as 33% of the highway maintenance budgets in some states (Hanbali, 1994). In Canada, 50% of total maintenance budget accounts for WRM (Buchanan & Gwartz, 2005), while expenditures for WRM in Japan and Europe can be even higher than those in the United States (Decker & Hanna, 1995; Hoffmann, Nutz, & Blab, 2012). Moreover, the most severe 20% of the storms can consume up to 80% of WRM budget (Shi, Huang, Williams, Akin, & Veneziano, 2014). Second, implication for countries in transition in SEE is the importance transportation has in their further development. In this case, road infrastructure plays a critical role, as it is essential to maintaining quality of life and economic growth (Croope & McNeil, 2007). One could even argue that there is no successful economic development without high-quality road network, just as there is no high-quality network without successful economic development (Kuzovic, Laisic, Glavic, Ostojic, & Pesic, 2009). Consequently, disruptions of traffic can have cascading effects on SEE countries’ economies, thus emphasizing the need for adaptation in WRM practices due to climate change.

Figure 1.

The region of South-East Europe.

The research presented here focuses on Serbia, as an exemplary country for SEE. First, considering a lack of transportation-related research about this particular region, this article will briefly describe a context of road transportation infrastructure in Serbia. Second, this article will present an overview of current WRM practices in Serbia, while drawing parallels from a large amount of previous international experiences. The comparison with the international WRM practice helps identify the core issues with WRM practice in Serbia. Third, this article will present a model for economic evaluation of WRM practice based on cost–benefit analysis (CBA), including a case study implementation. Finally, this article will draw conclusions about evaluation results followed by open issues for developing comprehensive WRM policies for SEE countries. Critical argumentation of existing practices, knowledge transfer from existing international experiences, and practice of economic evaluation used for justifying WRM investment decision should provide an excellent starting point for further improvement. Hence, this article should achieve one overall goal—providing a solid starting point for improvement of WRM policies in SEE region.

Transportation and Road Infrastructure Context in Serbia

The Republic of Serbia is one of the largest countries of SEE region. During the period 1990 to 2000, Serbia has been economically isolated, experiencing many economic discontinuities and low level of economic activity. Although having developed road network, this period has resulted in insufficient investments and inadequate maintenance. However, after political changes in 2000, the rehabilitation and development of the Serbian transport system was supported by international financial institutions, such as the European Investment Bank, the European Bank for Reconstruction and Development, and the World Bank. Currently, Serbia is experiencing rapid growth in transportation demand, with a massive shift of freight and passenger transport from rail to road transportation, accompanied with the increase in car ownership. As a result, road transportation in the Republic of Serbia represents a dominant mode of transport, with a share of approximately 80% in the total freight transportation and approximately 74% in total passenger transportation (“Strategy of Railway, Road, Inland Waterway, Air and Intermodal Transport Development in the Republic of Serbia, 2008–2015,” 2007).

Current road network with total length of 15,010 km is organized into four road categories. Category 1(a) has total length of 670 km, Category 1(b) has 4,100 km, Category 2(a) has 7,060 km, and Category 2(b) has 3,180 km. This road categorization is used for determining WRM priority category. The following map (Figure 2) presents road categorization and annual average daily traffic (AADT) on the highway network. One can observe that there are only several road sections where AADT is higher than 15,000 vehicles per day. The primary objective of Serbia’s road network is increasing accessibility within and between functional urban areas (“Spatial Plan of the Republic of Serbia 2010-2014-2020,” n.d.). Polycentric Western and Northern regions not only have higher network density but also have higher requirements from the perspective of demand. On the contrary, road network in less developed eastern and southern Serbia serves as a connection to the other regions of Serbia and SEE. Considering the existing road network and frequency of low transportation demand on primary road categories, maintenance investments have a priority over construction of new roads (Kuzovic, Laisic, Glavic, Vuksanovic, & Ostojic, 2009). With this in mind, and contrary to the common perception, Serbia and other SEE countries have existing WRM practices. However, as the next section will show, there are several issues with current WRM practices in Serbia.

Figure 2.

Road network categories and traffic volumes in Serbia.

Critical Overview of Current Winter Maintenance Practice in Serbia

This section identifies major features of WRM practice in Serbia. Considering that general WRM methodologies from different agencies are not directly transferable (Ye, Veneziano, Shi, & Fay, 2012), each point in this section first describes individual feature of WRM practice in Serbia. Afterward, for each point in this section, the article critiques elements of WRM practice in Serbia taking into account the latest international practice from developed countries.

In Serbia, the public enterprise “Roads of Serbia” (RS) has the primary role in WRM (“Strategy of Railway, Road, Inland Waterway, Air and Intermodal Transport Development in the Republic of Serbia, 2008–2015,” 2007). Municipal roads and street network are the responsibility of local government authorities. After the recent reform, public road maintenance bases on the system of fair tendering and contracting depending on the quality of maintenance services provided. This is one of the positive aspects of WRM in Serbia, as similar practice has been successfully implemented in many countries in the colder regions. Moreover, WRM period is fixed, lasting from November 15 to March 31. This fixed maintenance period does not take into account the fluctuation of climate, and appearance of frost outside of this period, especially during the fall.

WRM is organized through a network of headquarters and local bases, which receive information from the Republic Hydrometeorological Service of Serbia (national weather forecast service). This is a positive feature of WRM practice in Serbia, as international experience also recognizes the importance of weather information for WRM. However, similar to this international practice, weather forecasts in Serbia lack sufficient temporal and spatial details that can be used for location-based and real-time WRM measures (Pisano, 2001; Qiu & Nixon, 2009; Strong & Shi, 2008; Ye, Shi, Strong, & Greenfield, 2009). For example, this WRM practice neglects that temperature and wind speed at specific road sections can have significant effect on WRM effects (Qiu & Nixon, 2009). In addition, major drawback is that WRM activities are reactive, as they usually begin after the snowfall has decreased or stopped. Contrary to this, recent international practice has made a shift from reactive to proactive maintenance activities (Nordin & Arvidsson, 2014; Strong & Shi, 2008). For example, crucial components of recent international practice include change in criteria before WRM activities begin from 2 to 1 cm of snow (Nordin & Arvidsson, 2014) and use of anti-icing materials before the expected temperature change (Conger, 2005; Ketcham, Minsk, Blackburn, & Fleege, 1996; Nordin & Arvidsson, 2014; O’Keefe & Shi, 2005).

Roads in Serbia are classified into three WRM priority categories. WRM Priority 1 includes freeways and some sections of second-category roads. WRM Priority 1 or 2 includes all roads defined as Category 1(a) and Category 1(b). WRM Priority 2 or 3 includes all roads of Category 2. Similar practice of road prioritization for WRM is used worldwide. For example, in Canada, freeways, business areas, and transit/emergency routes are treated with priority (Nassiri, Bayat, & Salimi, 2014).

First, WRM priority roads are cleaned mechanically, and deiced using mixture of road salt and crushed stone aggregate. Second, WRM priority roads are cleaned mechanically, and deiced only using crushed stone aggregate. Third, WRM priority roads are only mechanically cleaned, after clearing roads of first priority and second priority. Moreover, a positive feature is that WRM procedure identifies critical points on the network due to, for example, high grade, sudden snow accumulation and icing, critical bridge location, difficulty in maintenance, longer periods of snow melting, and so on. However, WRM practice still uses a very simplistic procedure for assigning deicing and snow removal methods to specific road sections. One important, yet neglected, constraint is that road salt has reduced effectiveness in lower temperatures (Nassiri et al., 2014). Another neglected constraint is the environmental effect of road salt, which has even been prohibited in some locations (Mussato et al., 2007). Besides the environmental effects in the immediate road proximity, road salt can affect sanitary sewer water quality after washing maintenance equipment (Miller, Schneider, & Kennedy, 2015), and affect internal bridge corrosion (Jiang, Zheng, Liu, Wu, & Yang, 2011). Finally, holistic performance measurement system developed is not developed. Consequently, current practice does not document the level of service (LOS) provided, does not take into consideration a range of factors affecting WRM effectiveness, and does not assess the need for implementing different strategies (Brown, Perreault, & Baril, 2012; Ketcham et al., 1996; Pakkala, Jan, Lodenius, & Ohansson, 2011; Qiu, 2008).

During the winter maintenance period, staff and equipment are present 24 hr at the freeway bases, while 16 or 8 hr at bases on other roads. These staff assignments can also change in reaction to weather conditions. In addition, maintenance organizations are required to provide a minimum of 50% of the manpower, equipment, and deicing material for the highest level operation. For example, for 2014, predicted number of equipment used includes 701 transport vehicles and 308 snowplowing machines. In addition, there are 853 maintenance workers, 192 technicians, 310 machine operators, and 702 drivers. Cleaning machines are assigned to road sections based on altitude, road grades, wind strength, traffic volume, and general importance of specific road sections. In addition, reserves for deicing material are determined empirically, based on experience from previous years. For example, for WRM during 2013 and 2014, there has been stored a total of around 68,000 tons of road salt and 160,000 tons of crushed stone aggregate. These system elements are determined empirically, without any advanced scheduling operations or decision-support systems that could provide improved planning or real-time maintenance activities (Fay, Veneziano, Ye, Williams, & Shi, 2010; Ye, Strong, Shi, Conger, & Huft, 2009).

There are regulatory speed reductions during winter conditions. For first WRM priority roads, speed is limited to 60 km/hr, while for roads of second and third priority, speed limit is set at 40 km/hr. The use of this traffic regulation indicates obsolescent practice, where speed reduction consists of one predetermined value, being completely independent from the actual road condition. In addition, the use of winter driving equipment for vehicles (e.g., winter tires on all wheels, minimum tread depth for tires) is obligatory by law, which is a good practice adopted from Scandinavia.

Regulations accept traffic interruptions and delays in the duration of 1 or 2 hr, even for first-priority roads. Consequently, regulations for WRM practice accept externalizing costs to users. In practice, these interruptions can last longer, sometimes up to several days. Moreover, freight vehicles are sometimes forbidden to enter the country, thus kept waiting at the border crossing. However, at certain locations, traffic interruption for WRM activities is organized in cooperation with traffic police. This is a positive practice, as it can reduce the risk of accidents (Maze et al., 2005).

Information on road conditions is distributed to the public twice a day in regular weather conditions, or more frequently in high-precipitation weather conditions. Information is posted on RS website and sent to all major information agencies in Serbia. However, although planned (“Strategy of Railway, Road, Inland Waterway, Air and Intermodal Transport Development in the Republic of Serbia, 2008–2015,” 2007), there are no real-time information services available as in some countries in colder regions (Katko, 2006). Moreover, there are occasional campaigns for driving in winter conditions. On the contrary, there are no annual advertising campaigns or education about the tools for improved trip planning during winter (Brown et al., 2012; “Communication With Road Users in Winter Conditions,” 2013) or collection of feedback from travelers about road conditions (“Communication With Road Users in Winter Conditions,” 2013; Pakkala et al., 2011).

In overall, lack of well-developed and evaluated guidelines, reactive approach, static or fixed-temporal tactics primarily based on experience, lack of performance measurement procedures, externalization of costs to the user, and overemphasis of enforcement are just some of the major issues of WRM practice in Serbia. The general lack in understanding of WRM complexity and small significance assigned to WRM results in suboptimal resource allocation and insufficient financial investments. The research team has identified that the core of the problem is the lack of understanding of relationship between costs and benefits from WRM. For this reason, the next section of this article focuses on an exemplary case study for economic evaluation of investment in WRM.

Methodology for Winter Maintenance Investment Evaluation

The economic evaluation method selected for this study is CBA. In general, CBA is a process for calculating and comparing benefits and costs for a project, decision, or policy (“Notes on the Economic Evaluation of Transport Projects Transport Note No. TRN-5,” 2005). CBA has usually two goals:

To determine whether the investment or decision is economically or financially feasible and implementable;

To help in selecting an optimal solution and project comparison.

CBA can be used for both of these purposes, that is, to analyze economic feasibility of investing in WRM and to select certain WRM level. Based on the international experience, CBA analysis is one of the common methods for assessing the need for WRM investment (Boselly, 2001), despite some examples of other methods (e.g., microscopic traffic simulation; Shahdah & Fu, 2010). However, CBA is an unused method for evaluation of WRM investments in Serbia. As previously mentioned, the reason might be low understanding of costs and benefits related to WRM, accompanied with a disregard for economic evaluation of WRM investment decisions.

The parameters that CBA usually uses are internal rate of return (IRR), net present value (NPV), benefit over cost ratio (B/C), benefits (B), and costs (C). As the most suitable parameter for CBA of WRM, the B/C ratio parameter is selected for being best fitted to describe the ratio between investment in maintenance and return of savings. The methodology used a standard CBA, as opposed to an incremental CBA procedure. In the case of a standard CBA, B/C ratio compares benefits and costs for alternative WRM levels in contrast to the base scenario. In the case of an incremental CBA, B/C is obtained by comparing the differences in benefits and costs between alternative WRM levels, rather than base scenario. The choice of standard CBA procedure was influenced by the need for a broad argument, aiming to justify investments in WRM. This need for a broad argument originated from the identified low understanding of WRM costs and benefits among practitioners in Serbia.

Previous research identifies that the most common benefits of WRM are reduction of accidents and their severity, along with decrease in travel delay (Hanbali, 1994; Kuemmel & Hanbali, 1992; Sakshaug & Vaa, 1995). On the contrary, factors often associated with costs are material, labor, fuel consumption, vehicle corrosion, impact on environment, bridge corrosion, and so on. (Alppivuori et al., 1995; Maze et al., 2005; Wallman et al., 2006; Ye, Veneziano, & Shi, 2013). Many previous CBA studies have concluded the effectiveness of investment in winter maintenance. First among these studies, concluded that travel time costs (TTC) were reduced from 13.3 to 11.1 cents/vehicle mile traveled (VMT) and operational costs reduced from 8.6 to 7.8 cents/VMT (Kuemmel & Hanbali, 1992). In addition, total direct operating costs were reduced from 53.5 to 23.8 cents/VMT. Following study concluded a 20 cents worth of average direct savings to road user per vehicle kilometer of travel (Hanbali, 1994). Some recent studies have calculated B/C ratio being over 11.0 (Strong & Shi, 2008) and 6.0 (Ye et al., 2013). It is interesting to note that even for cases of complete snow removal with road heating, B/C ratio has ranged from 1.3 up to 9.2, depending on the region (Morisugi, Hayashiyama, Saito, & Sato, 2002).

For the purpose of this research, costs and benefits from WRM are divided as direct and indirect. Direct costs include material costs (e.g., road salt), equipment costs (e.g., fuel), and labor costs. Indirect costs are infrastructure costs (e.g., bridge corrosion, damage to the roadside equipment, etc.), vehicle costs (e.g., passenger vehicle corrosion from road salt), and environmental costs (e.g., water, vegetation or soil degradation from using deicing material). On the contrary, direct benefits include reduced TTC, benefits from reduced number of traffic accidents (i.e., reduced traffic accident costs [TAC]), reduced number of displaced trips (i.e., reduced displaced trips costs [IDC]), and reduction in fuel consumption and other vehicle operating costs (VOC). Moreover, indirect benefits include reduced number of compensation claims, continuance of industrial activity (e.g., reduced loss in production and freight delivery, reduced loss in wages, etc.), and enabling social activities (e.g., enhanced social interaction, connection between regions, enabling emergency vehicle response, etc.). Indirect benefits or costs can sometimes be much higher than direct benefits or costs, but accurate estimates for indirect benefits or costs are much more difficult to obtain (especially in the cases of traffic interruptions). In addition to determining the types of costs and benefits, it is of great importance to determine their economic value. The following sections will present a methodology for determining the value of benefits and costs from WRM (see Figure 3).

Figure 3.

A range of direct and indirect costs and benefits from winter maintenance.

Winter Roadway Maintenance Costs

Direct WRM costs are divided into three categories:

Material costs;

Labor costs; and

Equipment costs.

The model for quantifying WRM costs is (given through) the following function:

C_{{WM}_{k}} = f (A_{WM}, {WMS}_{k}, WWSI),

where $C_{{WM}_{k}}$ refers to costs for WRM practice at a level (k); $A_{WM}$ is the total area of maintenance (Section length × Road width); ${WMS}_{i}$ is the winter maintenance practice (k) that is a goal, and is often characterized with certain roadway condition index (RCI); and $WWSI$ is the winter weather severity index for the area.

For costs, empirical data have been collected from RS in Serbia and Public Enterprise Roads of Federation of Bosnia and Herzegovina in Bosnia and Herzegovina. These costs are used as examples of low WRM practice, as common throughout SEE region. In average, the cost of WRM on freeways in SEE region is 4,892 €/km. For WRM costs satisfying high WRM practice, empirical data from Germany have been used. In average, the costs of WRM on freeways in Germany during the period 2000 to 2010 were 6,930 €/km (Lippold, 2011).

Winter Roadway Maintenance Benefits

Direct WRM benefits are organized as follows:

Reduced TT;

Reduced number of traffic accidents;

Reduced VOC; and

Reduced number of displaced trips.

Method for estimating TT-related benefits

Benefits related to TT originate from reduced TT on the road section that has a certain level of WRM (k) compared with TT for the road section that has a low level of WRM (or even without WRM). Benefits from TT reduction are calculated as follows:

B_{{TT}_{k}} = L \times \sum_{i} [(\frac{1}{S_{NWM}} - \frac{1}{S_{{WM}_{k, i}}}) (n \times {ADT}_{i} \times O_{i} \times (\sum_{j} C_{VOT i_{j}} \times p_{i_{j}}))],

where $B_{{TT}_{k}}$ refers to benefits from TT savings for (k) level of WRM (€/winter season); $L$ is the road length (km); $S_{{WM}_{i}}$ is the travel speed for vehicle category (i) in conditions with no or low WRM (km/hr); $S_{{WM}_{k, i}}$ is the travel speed for vehicle category (i) in road conditions with (k) level of WRM (km/hr); $n$ is the number of days for driving in snowy conditions during the winter season (days); ${ADT}_{i}$ is the average daily traffic of vehicle category (i) during the winter season (veh/day); $O_{i}$ is the average occupancy of vehicles (passengers per vehicle); $C_{V O T_{i_{j}}}$ is the value of passenger TT in vehicle category (i) with the purpose of travel (j; €/h); and $p_{i_{j}}$ is the percentage of passengers in vehicle category (i) with purpose of travel (j).

In the case of TT costs, it is important to note that expected delays are valued at the rate of generalized costs, but unexpected delays are valued much higher as they result in interruption of synchronous activities at the trip’s destination (Maze et al., 2005).

Method for estimating benefits related to traffic accidents

For estimating WRM benefits related to traffic safety, it is important to determine the relationship between the number and severity of traffic accidents in different WRM conditions. Model for estimating benefits related to traffic accidents is a function of average value of traffic accident, and the number of traffic accidents per kilometer in cases with WRM at level (k), and with very low level of WRM (or even without WRM). The model is represented in the following equation:

B_{{TA}_{k}} = L \times C_{T A} \times (N_{NWM} - N_{{WM}_{k}}),

where $B_{{TA}_{k}}$ refers to benefits from traffic accidents savings for WRM at a level (k; €/winter season); $L$ is the road length (km); $C_{TA}$ is the average cost of one traffic accident, including all types of traffic accidents (€/traffic accident); $N_{NWM}$ is the number of traffic accidents per kilometer per winter season in conditions with no or low level of WRM; and $N_{{WM}_{k}}$ is the number of traffic accidents per kilometer per winter season in conditions with WRM at a level (k).

Method for estimating benefits related to VOC

Benefits related to VOC are obtained from the difference in VOC for vehicles traveling during k-level maintained and very low or unmaintained winter roadway conditions. The model is represented in the following equation:

B_{{VOC}_{k}} = L \times n (\sum_{i} {ADT}_{i} \times ({VOC}_{{NWM}_{i}} - {VOC}_{{WM}_{i, k}})),

where $B_{{VOC}_{k}}$ refers to benefits from VOC savings for at (k) level of WRM (€/winter season); $L$ is the road length (km); $n$ is the number of days for driving in winter conditions during the winter season (days); ${ADT}_{i}$ is the average daily traffic of vehicle category (i) during the winter season (veh/day); ${VOC}_{{NWM}_{i}}$ is the vehicle operating costs for vehicle category (i) in conditions without WRM or with low level of WRM (€/km); and ${VOC}_{{WM}_{i, k}}$ is the vehicle operating costs for vehicle category (i) in conditions with (k) level of WRM (€/km).

Method for estimating benefits related to displaced trips

Displaced trips include trips that are postponed or canceled, and consequently have a loss of productivity and loss of income (Fu, Usman, Miranda-Moreno, Perchanok, & McClintock, 2012). Benefits from reduced number of displaced trips are a function of average price for one displaced trip and the total number of displaced trips. The model is represented as follows:

B_{{IT}_{k}} = n \times {ADT}_{i} \times C_{IT} \times (p_{{IT}_{NWM}} - p_{{IT}_{WMk}}),

where $B_{{IT}_{k}}$ refers to benefits from displaced trips for WRM at level (k; €/winter season); $n$ is the number of days for driving in winter conditions during the winter season (days); ${ADT}_{i}$ is the average daily traffic of vehicle category (i) during the winter season (veh/day); $C_{IT}$ refers to costs for one displaced trip (€/trip); $p_{I T_{N W M}}$ is the percentage of displaced trips (i) in conditions without or with low level of WRM; and $p_{I T_{W M K}}$ is the percentage of displaced trips (i) in conditions with WRM on level (k).

Definition of Economic Evaluation Scenarios

For the purpose of economic evaluation for different WRM levels, the proposed evaluation parameter is B/C. To obtain B/C ratio for WRM practice on level (k), benefits and cost for each level (k) need to be determined. Three WRM levels are described with the following three scenarios:

Scenario 1 or base case: This scenario assumes no or very low level of WRM activities is carried out. This base scenario (k = 1) results in low traffic speed condition and very low LOS, resulting in significantly high TTC, VOC, TAC, and IDC.

Scenario 2 or present case of WRM in SEE region: This scenario (k = 2) corresponds to the actual WRM activities in Serbia, Bosnia and Herzegovina, FYR Macedonia, and Montenegro, where WRM activities are performed but at unsatisfactory level. Consequently, WRM Scenario 2 has unsatisfying traffic speed conditions and low LOS, resulting in high TTC, VOC, TAC, and IDC.

Scenario 3 or desired WRM in SEE region: This scenario (k = 3) corresponds to the actual (high level) WRM activities in colder region countries. In this case, an exemplary practice and empirical WRM cost data are taken from Germany, as country with high level of WRM closest to SEE region, and as countries in SEE region often use German standards and guidelines. WRM Scenario 3 has high traffic speed and high LOS during winter conditions, resulting in usual and acceptable TTC, VOC, TAC, and IDC.

Economic evaluation bases on comparing total costs between scenarios. For example, by comparing total costs from Scenario 2 with total costs from base Scenario 1, benefits can be determined as difference between these costs, thus presenting economic savings, that is, benefits. In addition, by comparing total costs from Scenario 3 with total costs from Scenario 1, one can obtain benefits for desired WRM level. As a result, decision makers can base their WRM investment decisions based on the obtained benefits from a certain WRM level. Next section will present a case study, implementing the proposed methodology.

Case Study

As an example of WRM economic evaluation process, this case study will focus on a freeway section in Serbia. The case study uses actual and empirically collected data.

Road Section Description and Evaluation Parameters

Motorway Belgrade–Novi Sad is situated in northern Serbia. Freeway has two traffic lanes per direction, with an emergency lane, and 4 m wide median. In addition, freeway has full access control, with all intersections constructed as interchanges. AADT of this section is around 13,000 vehicles per day. The location of the freeway section is presented on Figure 4.

Figure 4.

Location of motorway section on road map of Serbia.

Input data for CBA evaluation are presented in Table 1. These data have been obtained from roadway designs, calculation using software HDM (“HDM-4 The Highway Development and Management Series,” 2000), and various empirical sources.

Table 1.

Input Data for WRM CBA.

Parameter	Value	Source
ADT	13,500 veh/day	Automatic counters
L	80 km	Final roadway design
$S_{{NWM}_{i}}$	60 km/hr	Empirical data
$S_{{WM}_{k, i}}$	$S_{W M_{k = 2}}$ = 80 km/hr; $S_{W M_{k = 3}}$ =100 km/hr	Empirical data
$n$	42.6	Meteorological agency
$C_{IT}$	10.0 €	Estimate
$p_{I T_{k}}$	$p_{IT k = 1} = 0.20;$ $p_{IT k = 2} = 0.10; p_{IT k = 3} = 0.00$	Estimate
${VOC}_{{NWM}_{i}}$ ${VOC}_{{WM}_{i, k}}$	${VOC}_{{NWM}_{PC}} = 0.20 € / km;$ ${VOC}_{{NWM}_{HT&BUS}} = 1.10 € / km$ ${VOC}_{{NWM}_{{PC}_{K = 2}}} = 0.18 € / km;$ ${VOC}_{{NWM}_{{HT&BUS}_{K = 2}}} = 1.05 € / km$ ${VOC}_{{NWM}_{{PC}_{K = 3}}} = 0.17 € / km;$ ${VOC}_{{NWM}_{{HT&BUS}_{K = 3}}} = 0.95 € / km$	HDM-4
$C_{TA}$	10,600 €	Empirical data
$N_{NWM}, N_{{WM}_{k}}$ ,	$N_{NWM} = \frac{0, 8 TA}{km}; N_{{WM}_{k = 2}} = \frac{0.4 TA}{km}; N_{{WM}_{k = 2}} = \frac{0.3 TA}{km}$	Statistical and analytic data
$O_{i}$	$O_{PC} = 2.1; O_{BUS} = 45$	Empirical data used in feasibility studies in region
$C_{i_{j}}$	$C_{{PC}_{work}} = 3.0 €; C_{{PC}_{other}} = 1.0 €;$ $C_{{BUS}_{work}} = 2.7 €; C_{{PC}_{other}} = 0.9 €$	Empirical data based on average annual income in Serbia
$P_{i_{j}}$	$p_{{PC}_{work}} = 25 %; p_{{PC}_{other}} = 75 %;$ $C_{{BUS}_{work}} = 33 %; C_{{PC}_{other}} = 67 %$	Empirical data used in feasibility studies in region

Note. WRM = winter road maintenance; CBA = cost–benefit analysis; ADT = average daily traffic.

Results and Discussion

Based on the data for this section, and applying the defined methodology through Equations 1 to 5, the following economic evaluation parameters were determined (Table 2.)

Table 2.

WRM CBA Results Data.

Parameter	Scenario 2SEE WRM standard	Scenario 3Highest WRM standard
$B_{VOC}$ (€)	898,128	1,903,306
$B_{T T_{k}}$ (€)	737,100	1,179,360
$B_{TA}$ (€)	636,000	1,017,600
$B_{IT}$ (€)	283,500	850,500
Total benefits (€)	2,554,728	4,950,766
WRM costs (€)	391,385	554,400
B/C ratio *	6.5	8.9

Note. WRM = winter road maintenance; CBA = cost–benefit analysis; SEE = South-East Europe; B/C = benefit over cost ratio.

Bolface indicates the key outcomes of the CBA and final values of B/C ratio.

The value of B/C parameter greater than one implies that investing in certain action is economically justified. As one can observe from Table 2, B/C ratio for both Scenarios 2 and 3 is significantly higher than one. This fact leads to a conclusion that it is economically justifiable to invest in WRM in SEE region. For every monetary unit invested, benefits could be 6 to 9 times greater. However, if decision makers do not invest in WRM, they will create substantial economic, as well as, social losses. In addition, if decision makers decide to invest in higher WRM level, direct benefits will be even greater. This is an excellent argument for investing in higher WRM level, similar to developed countries in colder regions. Here, one should note that if the calculated B/C ratio for lower WRM level has been lower than B/C ratio for higher WRM level, an incremental CBA procedure should be used. In this case, incremental CBA would precisely determine economic justification of investing in WRM.

In practice, decision makers in SEE region often focus solely on WRM costs. This approach often neglects the fact proven here, that with higher investments in WRM, and consequent higher WRM level obtained, the decision is completely justified from the economic standpoint. Moreover, the justification of investments in WRM would be even greater if indirect benefits are taken into account. Results obtained from this case study can be expanded to the rest of the road network in the region, as B/C is universal non-dimensional parameter. This generalization is supported by the fact that even in the case traffic volume for specific road section is 1,000 vehicles per day (which is not the case for any freeway section in the region), B/C ratio does not fall below one. Conclusively, if transportation agencies in the region are to promote sustainable practices, they need to consider an important argument of economic efficiency (Haynes, Gifford, & Pelletiere, 2005). From the standpoint of economic efficiency, and considering the need for quality transportation service in the region, it is justified to invest in WRM.

The results from this case study show the applicability and importance of economic evaluation of WRM investment decisions. In addition, considering that this research effort intended to provide a solid starting point for improvement of WRM policies in SEE region, there are several points to consider.

A much-needed change in institutional culture needs to include knowledge transfer from existing state of the practice of agencies in the colder regions. The “Critical Overview of Current Winter Maintenance Practice in Serbia” section of this article has provided some key discrepancies between practices and starting points for knowledge transfer. However, it is important to emphasize one point. Considering the features of incoming climate change (especially infrequent but extreme storms accompanied with low temperatures and precipitation), the solution is not simply to increase existing maintenance capacities, but also to extend the range of strategies and tactics.

Transportation agencies in SEE region need to improve their performance measurement systems. Analytically based performance criteria (Neumann & Markow, 2004) tied to agency’s policy objectives (Haas, Felio, Lounis, & Falls, 2009) can significantly improve investment decisions. An excellent starting point is existing international experiences of quantitative and qualitative indicators for WRM performance measurement and evaluation.

Improvements in WRM practice might not require significant monetary investments, but rather standardization across agencies, and improvements in WRM methods. Furthermore, comparatively small investments in Intelligent Transport System technologies for weather information and data collection can improve decision making for both transportation agencies and users. For example, by implementing web-based roadway weather information systems and variable message signs at critical locations, speed recommendations could depend on RCI, as opposed to having single predetermined values based on road category.

Conclusion

This article started with the argument that weather, as important factor for traffic and roadway conditions, will become even more important for SEE countries due to the evident climate change. On the contrary, countries in transition in SEE region are highly dependent on their roadway infrastructure for their further economic development. Within this climate and economic context, winter roadway maintenance practices could have an important role for enabling normal traffic conditions, and consequently enabling further economic development. Contrary to some common misconceptions, WRM practices exist in SEE countries. However, these WRM practices, as exemplified on the case of WRM practice in Serbia, are far from the level of WRM practices in developed countries in the colder regions. WRM practice in Serbia lacks holistic performance measurement system that improves assessment of implementing different strategies; relies on reactive, static, and empirical tactics; externalizes costs to the user; and emphasizes enforcement.

The core of the problem is lack of understanding of costs and benefits from WRM. Moreover, the reason for lacking investment in WRM might be that transportation professionals in SEE region do not completely appreciate the importance of economic evaluation of WRM investment decisions. However, as the methodology and case study presented here have shown, it significantly pays off to invest in WRM, even on a higher level than current WRM practice. Consequently, if transportation agencies in the region are to promote sustainable practices for quality transportation service, the important argument of economic efficiency is justifying investments in WRM. In addition to several recommendations for improving WRM practice, there is one final point to emphasize. The improvement of the overall process of WRM investment decisions needs to start with a change in institutional culture. An excellent starting point for this change is a widespread introduction of economic evaluation of WRM investment decisions.

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

Author Biographies

Drazenko Glavic is an assistant professor at the Department for Traffic Engineering at University of Belgrade in Serbia. He is a leading South East Europe expert in transport economy, feasibility studies, and traffic analysis, with 18 years of experience and over 120 studies.

Milos N. Mladenovic is an assistant professor at Department of Civil and Environmental Engineering at Aalto University in Finland. His research areas include transport systems planning, transport infrastructure management, and transport modeling.

Aleksandar Stevanovic is an associate professor of Civil, Environmental and Geomatics Engineering at Florida Atlantic University, and a licensed Professional Engineer in the state of Utah. His main research areas include traffic operations, intelligent transportation systems, and infrastructure evaluation.

References

Alppivuori

Leppänen

Anila

Mäkelä

(1995). Road traffic in winter. (Finnra Reports, 57/1995). Helsinki, FI: Finnish National Road Administration (Finnra), Traffic Services.

Andersson

A. K.

Chapman

(2011). The impact of climate change on winter road maintenance and traffic accidents in West Midlands, UK. Accident Analysis & Prevention, 43, 284-289. doi:10.1016/j.aap.2010.08.025

Andreescu

M.-P.

Frost

D. B.

(1998). Weather and traffic accidents in Montreal, Canada. Climate Research, 9, 225-230.

Belgrade initiative: Enhancing the regional SEE cooperation in the field of climate change—Climate change framework action plan for the SEE region, and the establishment of a sub-regional, virtual climate change related centre for research and systematic observation, education, training, public awareness, and capacity building. (2007). In Sixth Ministerial Conference “Environment for Europe.” Belgrade, Serbia. Retrieved from http://www.unece.org/fileadmin/DAM/env/documents/2007/ece/ece.belgrade.conf.2007.20.add1.e.pdf

Berkhout

Mechler

Neufeldt

Patt

Watkiss

Wreford

. . . Egenhofer

(2009). Adaptation to climate change: Why is it needed and how can it be implemented? (CEPS Policy Brief No. 161). Brussels, Belgium: Centre for European Policy Studies.

Bernstein

Bosch

Canziani

Chen

Christ

Davidson

. . . Yohe

(2007). Climate change 2007: An assessment of the intergovernmental panel on climate change. Geneva, Switzerland: Intergovernmental Panel on Climate Change.

Boselly

(2001). Benefit/cost study of RWIS and anti-icing technologies (Vol. 7). Washington, DC: Final report prepared for National Cooperative Highway Research Program, Transportation Research Board, National Research Council.

Brown

Perreault

Baril

(2012). Winter road maintenance at the Ministère des Transports du Québec. Cold Regions Engineering, 2012, 308-318. doi:10.1061/9780784412473.031

Buchanan

Gwartz

(2005). Road weather information systems at the Ministry of Transportation, Ontario. Annual Conference of the Transportation Association of Canada, Calgary. Retrieved from http://trid.trb.org/view.aspx?id=798228

10.

Cao

Thakali

Donaher

(2013). Effect of weather and road surface conditions on traffic speed of rural highways. In Transportation Research Board 92nd Annual Meeting (Vol. 28). Washington, D.C. Retrieved from http://docs.trb.org/prp/13-0779.pdf

11.

Communication with road users in winter conditions. (2013). Retrieved from http://trid.trb.org/view.aspx?id=1247050

12.

Conger

S. M.

(2005). NCHRP synthesis 344—Winter highway operations: A synthesis of highway practice. Washington, DC: Transportation Research Board.

13.

Croope

McNeil

(2007). Understanding critical infrastructure-management challenges for the 21st century. 11th World Conference on Transport Research, Berkeley, CA. Retrieved from http://trid.trb.org/view.aspx?id=883830

14.

Decker

Hanna

A. N.

(1995). Winter maintenance technology and practices—Learning from abroad (Vol. 204). NCHRP Research Results Digest Number, Washington, DC. Retrieved from http://pubsindex.trb.org/view/1995/m/457853

15.

Eisenberg

(2004). The mixed effects of precipitation on traffic crashes. Accident Analysis & Prevention, 36, 637-647. doi:10.1016/S0001-4575(03)00085-X

16.

Eisenberg

Warner

K. E.

(2005). Effects of snowfalls on motor vehicle collisions, injuries, and fatalities. American Journal of Public Health, 95, 120-124. doi:10.2105/AJPH.2004.048926

17.

Enberg

Mannan

(1998). Effects of road and weather conditions on traffic flow on a three-lane rural highway in Finland. In Xth PIARC International Winter Road Congress. Luleå, Sweden. Retrieved from http://trid.trb.org/view.aspx?id=488018

18.

Fay

Veneziano

D. A.

Williams

Shi

(2010). Costs and benefits of tools to maintain winter roads: A renewed perspective based on recent research. Transportation Research Record: Journal of the Transportation Research Board, 2169, 174-186. doi:10.3141/2169-19

19.

Usman

Miranda-Moreno

L. F.

Perchanok

M. S.

McClintock

(2012). How much is the safety and mobility benefit of winter road maintenance? Retrieved from http://conf.tac-atc.ca/english/annualconference/tac2012/docs/session7/usman.pdf

20.

Gocic

Trajkovic

(2013). Analysis of changes in meteorological variables using Mann-Kendall and Sen’s slope estimator statistical tests in Serbia. Global and Planetary Change, 100, 172-182. doi:10.1016/j.gloplacha.2012.10.014

21.

Haas

Felio

Lounis

Falls

L. C.

(2009). Measurable performance indicators for roads: Canadian and international practice. In 2009 Annual Conference of the Transportation Association of Canada. Vancouver, British Columbia, Canada.

22.

Hanbali

R. M.

(1994). Economic impact of winter road maintenance on road users. Transportation Research Record: Journal of the Transportation Research Board, 1442, 151-161. Retrieved from http://trid.trb.org/view.aspx?id=414444

23.

Haynes

K. E.

Gifford

J. L.

Pelletiere

(2005). Sustainable transportation institutions and regional evolution: Global and local perspectives. Journal of Transport Geography, 13, 207-221. doi:10.1016/j.jtrangeo.2004.07.003

24.

HDM-4 the highway development and management series. (2000). University of Birmingham. Retrieved from http://www.piarc.org/en/knowledge-base/road-assets-management/HDM-4-Software/

25.

Hoffmann

Nutz

Blab

(2012). Dynamic modeling of winter maintenance strategies and their impact on skid resistance. Procedia—Social and Behavioral Sciences, 48, 682-691. doi:10.1016/j.sbspro.2012.06.1046

26.

Jiang

Zheng

Liu

Yang

(2011). Studies review of the technology for snow and ice control for winter road maintenance. In Third International Conference on Transportation Engineering (ICTE) (pp. 3245-3254). Retrieved from http://ascelibrary.org/doi/abs/10.1061/41184%28419%29535

27.

Katko

T. S.

(2006). Road safety fatalities, management, and policy in Finland 1970-2003,. Public Works Management & Policy, 11, 126-138. doi:10.1177/1087724X06294066

28.

Keay

Simmonds

(2006). Road accidents and rainfall in a large Australian city. Accident Analysis & Prevention, 38, 445-454. doi:10.1016/j.aap.2005.06.025

29.

Ketcham

S. A.

Minsk

L. D.

Blackburn

Fleege

(1996). Manual of practice for an effective anti-icing program. Federal Highway Administration FHWA-RD-95-202. Washington, DC.

30.

Koetse

M. J.

Rietveld

(2009). The impact of climate change and weather on transport: An overview of empirical findings. Transportation Research Part D: Transport and Environment, 14, 205-221. doi:10.1016/j.trd.2008.12.004

31.

Kuemmel

D. A.

Hanbali

R. M.

(1992). Accident analysis of ice control operations. The Salt Institute. Alexandria, Virginia, USA.

32.

Kuzovic

Laisic

Glavic

Ostojic

Pesic

(2009). Infrastructure and economic development—Hen or egg question. In 16th World Meeting of International Road Federation. Lisbon. Portugal.

33.

Kuzovic

Laisic

Glavic

Vuksanovic

Ostojic

(2009). Sustainable development of Serbian national roads. In 16th World Meeting of International Road Federation. Lisbon, Portugal.

34.

Kwon

T. J.

Jiang

(2013). Effect of winter weather and road surface conditions on macroscopic traffic parameters. Transportation Research Record: Journal of the Transportation Research Board, 2329, 54-62.

35.

Lippold

(2011). Der Elsner-Handbuch für Strassen und Verkehrswesen. [Der Elsner - Manual for Roads and Transportation]. Berlin, Germany: Otto Elsner Verlagsgesellschaft.

36.

Maze

T. H.

Agarwai

Burchett

(2006). Whether weather matters to traffic demand, traffic safety, and traffic operations and flow. Transportation Research Record: Journal of the Transportation Research Board, 1948, 170-176. Retrieved from http://trb.metapress.com/content/d228243908410567/

37.

Maze

T. H.

Crum

M. R.

Burchett

(2005). An investigation of user costs and benefits of winter road closures (InTrans Project Reports, Paper 21). Retrieved from http://lib.dr.iastate.edu/cgi/viewcontent.cgi?article=1019&context=intrans_reports

38.

Miller

C. M.

Schneider

W. H.

Kennedy

M. J.

(2015). Evaluation of management alternatives for truck wash water generated during winter maintenance activities. Public Works Management & Policy, 20, 80-100. doi:10.1177/1087724X14522601

39.

Minsk

D. L.

(1998). Snow and ice control manual for transportation facilities. New York, NY: McGraw-Hill.

40.

Morisugi

Hayashiyama

Saito

Sato

(2002). Benefit evaluation of road snow removal in the Tohoku region. In New Challenges for Winter Road Service. XIth International Winter Road Congress (Vol. d, pp. 1-10). Retrieved from http://www.econ.tohoku.ac.jp/~yhaya/II118e.pdf

41.

Mussato

B. T.

Gepraegs

O. K.

Seabrook

P. T.

Davison

J. P.

Charlton

R. S.

Parker

R. D.

. . . Chollar

(2007). NCHRP report 577: Guidelines for the selection of snow and ice control materials to mitigate environmental impacts. Washington, DC: Transportation Research Board.

42.

Nassiri

Bayat

Salimi

(2014). Survey of practice and literature review on municipal road winter maintenance in Canada. Journal of Cold Regions Engineering. Retrieved from http://ascelibrary.org/doi/abs/10.1061/(ASCE)CR.1943-5495.0000082

43.

Neumann

L. A.

Markow

M. J.

(2004). Performance-based planning and asset management. Public Works Management & Policy, 8, 156-161. doi:10.1177/1087724X03259474

44.

Nordin

Arvidsson

A. K.

(2014). Are winter road maintenance practices energy efficient? A geographical analysis in terms of traffic energy use. Journal of Transport Geography, 41, 163-174. doi:10.1016/j.jtrangeo.2014.09.004

45.

Notes on the economic evaluation of transport projects transport note no. TRN-5. (2005). The World Bank. Washington, DC.

46.

O’Keefe

Shi

(2005). Synthesis of information on anti-icing and pre-wetting for winter highway maintenance practices in North America. Bozeman, MT: Pacific Northwest Snowfighters Association & Washington State Department of Transportation. Retrieved from http://www.westerntransportationinstitute.org/documents/reports/4w0169_final_report.pdf

47.

Pakkala

Jan

Lodenius

Ohansson

A. W.

(2011). Sweden’s “the road to excellence” international benchmarking study of customer satisfaction for winter services and pavements. In 90th Annual Meeting of Transportation Research Board, Washington, DC. Retrieved from http://docs.trb.org/prp/11-0213.pdf

48.

Pisano

(2001). Foretell—Finally, someone is doing something about the weather! Public Roads, 64(5), 15-16.

49.

Qiu

(2008). Performance measurement for highway winter maintenance operations. Ann Arbor, MI: ProQuest.

50.

Qiu

Nixon

(2009). Performance measurement for highway winter maintenance operations (IIHR Technical Report No. 474). Iowa City, IA: Iowa Highway Research Board. Retrieved from http://www.iihr.uiowa.edu/wp-content/uploads/2013/06/IIHR474.pdf

51.

Sabir

van Ommeren

Koetse

Rietveld

(2008). Welfare effects of adverse weather through speed changes in car commuting trips (Vol. 41). Amsterdam, Netherlands: Tinbergen Institute.

52.

Sakshaug

Vaa

(1995). Road salting and traffic safety (SINTEF transport engineering, Report No. STF63). Trondheim, Norway: SINTEF.

53.

Shahdah

(2010). Quantifying the mobility benefits of winter road maintenance—A simulation based analysis. In TRB 89th Annual Meeting Compendium of Papers (Vol. 33984), Washington, DC.

54.

Shi

Huang

Williams

Akin

Veneziano

(2014). Highway winter maintenance operations at extremely cold temperatures. In Proceedings of the 2014 International Symposium on Climatic Effects on Pavement and Geotechnical Infrastructure (pp. 52-65).

55.

South East European climate change framework—Executive summary and action plan. (2008). Retrieved from http://documents.rec.org/topic-areas/CCFAP_Executive_Summary_final.pdf

56.

Spatial plan of the republic of Serbia 2010-2014-2020. (n.d.). Retrieved from http://195.222.96.93//media/zakoni/Spatial%20Plan%20of%20the%20Republic%20of%20Serbia_2010-2020_abridged%20(1).pdf

57.

Strategy of railway, road, inland waterway, air and intermodal transport development in the Republic of Serbia, 2008–2015. (2007). Retrieved from http://www.putevi-srbije.rs/strategijapdf/Strategijatransport_eng.pdf

58.

Strong

Shi

(2008). Benefit-cost analysis of weather information for winter maintenance: A case study. Transportation Research Record: Journal of the Transportation Research Board, 2055, 119-127.

59.

Tsapakis

Cheng

Bolbol

(2013). Impact of weather conditions on macroscopic urban travel times. Journal of Transport Geography, 28, 204-211. doi:10.1016/j.jtrangeo.2012.11.003

60.

Unkašević

Tošić

(2008). Changes in extreme daily winter and summer temperatures in Belgrade. Theoretical and Applied Climatology, 95, 27-38. doi:10.1007/s00704-007-0364-7

61.

Unkašević

Tošić

(2013). Trends in temperature indices over Serbia: Relationships to large-scale circulation patterns. International Journal of Climatology, 33, 3152-3161. doi:10.1002/joc.3652

62.

Usman

Miranda-Moreno

L. F.

(2012). A disaggregate model for quantifying the safety effects of winter road maintenance activities at an operational level. Accident Analysis & Prevention, 48, 368-78. doi:10.1016/j.aap.2012.02.005

63.

Wallman

C.-G.

Möller

Blomqvist

Gustafsson

Niska

Öberg

. . . Karlsson

B. O.

(2006). The winter model, stage 2 (Final report) (VTI rapport 531). Linköping, Sweden: Swedish National Road and Transport Research Institute.

64.

Wallman

C.-G.

Wretling

Öberg

(1997). Effects of winter road: State-of-the-art (VTI rapport 423A). Linköping, Sweden: Swedish National Road and Transport Research Institute.

65.

Williamson

S. P.

Estis

F. L.

(2007). Weather Information for Surface Transportation (WIST): Update on weather impacts and WIST progress. In Proceedings of the 2007 Mid-Continent Transportation Research Symposium.

66.

Shi

Strong

C. K.

Greenfield

T. H.

(2009). Evaluation of effects of weather information on winter maintenance costs. Transportation Research Record: Journal of the Transportation Research Board, 2107, 104-110. doi:10.3141/2107-11

67.

Strong

C. K.

Shi

Conger

S. M.

Huft

D. L.

(2009). Benefit-cost analysis of maintenance decision support system. Transportation Research Record: Journal of the Transportation Research Board, 2107, 95-103. doi:10.3141/2107-10

68.

Veneziano

Shi

(2013). Estimating statewide benefits of winter maintenance operations. Transportation Research Record: Journal of the Transportation Research Board, 2329, 17-23.

69.

Veneziano

Shi

Fay

(2013). Estimating statewide benefits of winter maintenance operations. Transportation Research Record: Journal of the Transportation Research Board, 2329, 17-23. doi:10.3141/2329-03.