Richard J. Porter Assistant Professor Department of Civil and Environmental Engineering, University of Utah 110 Central Campus Dr., Rm.

Transcription

1 Street Connectivity vs. Street Widening: Impact of Enhanced Street Connectivity on Traffic Operations in Transit Supportive Environments Ivana Tasic* Graduate Research Assistant Civil and Environmental Engineering, University of Utah 0 Central Campus Dr., Rm. 0 Salt Lake City, Utah Phone: (0) - Fax: (0) -0 ivanat@trafficlab.utah.edu *Corresponding Author Milan Zlatkovic Research Assistant Professor Civil and Environmental Engineering, University of Utah 0 Central Campus Dr., Rm. 0 Salt Lake City, Utah Phone: (0) - Fax: (0) -0 Milan@trafficlab.utah.edu Peter T. Martin Department of Civil Engineering, New Mexico State University P.O. Box 000, Hernandez Hall Room 0 Las Cruces, NM Phone : () -0 Fax : () -0 wales@nmsu.edu Richard J. Porter Assistant Professor Department of Civil and Environmental Engineering, University of Utah 0 Central Campus Dr., Rm. 000 Salt Lake City, Utah Phone: (0) -0 Fax: (0) - richard.jon.porter@utah.edu Word Count: + ( Figures + Tables) = Prepared for the Transportation Research Board, 0 November, 0

2 I. Tasic, and M. Zlatkovic 0 0 ABSTRACT Highly connected street networks increase accessibility for multimodal transport, but their effects on the efficiency of still dominant vehicular traffic is rarely addressed. As the interest in retrofitting the typical suburban developments from car-oriented to multimodal environments increases, the effects of redesigned street networks in the period before the expected mode shift occurs need to be clarified. This paper addresses the implications of enhanced connectivity on traffic operations on the part of the West Valley City network in Utah, as the potential Transit Oriented Development (TOD) site. Since predicted traffic demand for 00 requires modifications of this network, the question is if enhanced connectivity, as a TOD supportive approach, can accommodate that demand and replace the traditional street widening solution. A total of twelve scenarios were modeled and evaluated including: the existing state, five scenarios with different levels of street connectivity, five street widening scenarios, and an additional scenario with reduced speed areas based on the traffic calming practice. Macrosimulation and microsimulation models were used iteratively to build, calibrate and evaluate the modeled scenarios. The results on the intersection, corridor and network-wide level show that enhanced street connectivity represents a competitive alternative for the traditional capacity expansion approaches that usually involve street widening. As connectivity increases, network designs with enhanced connectivity accommodate more traffic than designs with street widening, opening new routes and providing a better dispersion of intra-zonal traffic. New scenarios that encompass changes in mode split are proposed for the future research efforts.

3 I. Tasic, and M. Zlatkovic INTRODUCTION Many of the existing transportation issues related to public health and the environment are attributed to urban automobile dependency, and one of the causes of automobile dependency is often argued to be street network pattern with the low level of street connectivity. Street connectivity is a topic on which engineers and planners both agree and disagree, but it is a very current topic in both research and practice. Urban planners are usually proponents of increased street connectivity, while engineers sometimes hesitate to implement solutions that include higher levels of street connectivity. This is due to the fact that the existing research does not consistently explain or quantify the benefits of street connectivity for transportation system efficiency. The term street connectivity refers to the original purpose of streets which is to connect places and enable safe and efficient movement of people to their destinations (). The potential benefits of street connectivity are often mentioned in the literature, but rarely quantified in terms of traditional transportation performance measures. The existing research rarely uses actual transportation performance measures to communicate the findings on the impacts of street connectivity. The possibility that only a certain amount of connectivity is optimal, and that there is a need to balance the density of the street networks in urban environments is rarely examined. This is why a potentially beneficial concept of well-connected streets is sometimes implemented with hesitation in long term transportation plans, although it is often proposed by urban land developers. The existing gaps in research, especially in terms of quantification methods, should be overcome in order to enable transportation planners to design safer, more efficient street networks with optimal amount of connectivity. Some recent land use and transportation planning efforts in Wasatch Front Region, especially Salt Lake City Downtown area, involve consideration of enhanced street connectivity, as a way to support multimodal transportation, without compromising the efficiency of traffic operations for vehicular traffic (). This includes the initiative for Walkable Salt Lake, where downtown midblock walkways project started to address walkability in the area, recognizing that Salt Lake City s large 0-acre blocks can be overwhelming to pedestrians (). The Sixty Nine Seventy competition of urban design ideas was held in 0, again with the purpose to revitalize and break large blocks in Salt Lake Downtown (). Salt Lake City Master Plan is a vision for the future years of development, which also addresses midblock streets and walkways (, ). As the number of these initiatives increases, the effect of street connectivity on traffic operations remains unexamined. The objectives of this paper are to identify research gaps in the existing literature that attempts to quantify the effects of street connectivity, and explore the need to balance street connectivity in the areas where long term transportation plans are set up to develop multimodal transportation systems. To achieve these objectives, the paper uses a case study network of the future Transit Oriented Development (TOD) spot in West Valley City, Utah (Figure ). The first section of this paper explains the general understanding of street connectivity and its benefits, as well as the local efforts to increase street connectivity in Wasatch Front Region, defining the research objectives. The second section is the discussion of current research approaches to quantifying the benefits of street connectivity. The third section explains the methodology, while the fourth section presents the results. The discussion and conclusions are provided in the final section of the paper.

4 I. Tasic, and M. Zlatkovic 0 FIGURE Study network with Utah Transit Authority (UTA) transit lines and stops. LITERATURE REVIEW Street connectivity indicates how well the street system connects different parts of an area (). Proponents of street connectivity argue that the quality of connections or the connectivity of the street network influences the accessibility of potential destinations and has important implications for travel choices, emergency access, and more generally, quality of life (). The Congress for New Urbanism is also promoting the concept of connectivity as a part of effort to create more livable and sustainable communities (). Higher connectivity is considered to be beneficial as it provides more direct routes to any destination and enables better traffic dispersion ().

5 I. Tasic, and M. Zlatkovic Street Connectivity Measures There are many studies that deal with the problem of measuring the street connectivity (-). One of the most common issues addressed in these studies is choosing the appropriate measure and method of measuring street connectivity (, ). Each connectivity measure links travel behavior to urban forms (, ). The purpose is to determine the standards and ranges of connectivity that would both benefit residential areas and not disrupt the flow of traffic for a variety of vehicle types (). Connectivity measures can be deployed as performance standards for new and/or existing development (,, -). The calculations of street connectivity measures may require a very detailed and complex approach (0, ). Some of the measures of street connectivity from the previous research include: block length, size and density (,, ); effective walking area and pedestrian catchment area (); intersection and street density (); connectivity index (); directional reach and global metric betweenness (). The application of Geographic Information Systems significantly facilitated connectivity measurement methods over the years (). Traffic Impacts of Street Connectivity Some of the previously reviewed street connectivity measures were utilized to establish the relationships between the level of street network connectivity and transportation related outcomes. Perhaps the greatest potential benefit of increasing street connectivity that is often mentioned in the literature is improving transportation safety (, ). One of the most comprehensive studies in this area used data on more than 0,000 crashes occurring over nine years in medium-sized California cities and evaluated against principal measures of street network connectivity. The results of this study suggest that street network characteristics do in fact play a role in road safety outcomes. However the underlying factors contributing to this role are not yet known (0). Another study that relates built environment to traffic safety, using some density indicators, concludes that denser urban areas could be considered safer due to lower exposure to long distance travel and lower driving speeds (). Another potential benefit of street connectivity from the available literature is a higher percentage of people walking, biking, and taking transit leading to lower vehicle emissions and healthier lifestyle (-). Both increased intersection density and additional street connectivity were generally associated with more walking, biking, and transit use, but the causal relationship was not always established (). The engineers could be concerned about effects of street connectivity on the efficiency of traffic in urban environments, asking what the immediate impact on vehicular traffic congestion is. Existing research states that highly connected networks of narrower streets have higher capacity than lower density networks with wider streets, and have the potential to reduce traveler s delay, but these studies rarely use actual measures of capacity to quantify this relationship (, ). Very few efforts have evaluated the relationships between street connectivity and performance indicators such as delay, capacity, and travel time, proving that connectivity can significantly affect traffic operations, and indicating the need for additional research in this area (, 0). While it is certain that enhanced street connectivity has the potential to affect mode split (-), and thus reduce the percent of people driving and potentially reduce congestion, it is still unclear what the immediate effects of street connectivity are, before the actual changes in mode split are achieved. This could particularly be an issue in suburban developments that attempt to convert from car-oriented to multimodal environments. This paper focuses on one such development in West Valley City, Utah, in order to evaluate the immediate effects of

6 I. Tasic, and M. Zlatkovic 0 enhanced street connectivity on traffic operations in the period of transition when the mode shift still did not occur. The case study network analyzed in this paper is the future TOD location, and exploring the transitional effects of transit supportive efforts such as enhanced street connectivity is beneficial for the eventual TOD implementation. METHODOLOGY This paper explores the effects of both increasing street network connectivity and the traditional street widening approach on the network traffic operations. Design principles for improving the way streets are connected are adapted from the reviewed literature (-). The approach used in this paper increases the connectivity between the streets in the study network gradually, until the recommended level of network connectedness is achieved. The modeling methodology process is presented in Figure. Simulation scenario Calibrated network OD matrix for time period VISUM Static Traffic Assignment Intersection turning volumes ANM Export Geometry tuning Synchro signal optimization Existing signal timings (i) VISSIM network Signal timings Performance measures (intersection, link, network level) FIGURE Modeling methodology.

7 I. Tasic, and M. Zlatkovic 0 0 The effects of the implemented network design principles were assessed through combined macro and micro traffic simulations. The models were developed simultaneously in VISUM (macro) and VISSIM (micro) simulation software. The main inputs used in the state of development and calibration of models were network geometry, traffic analysis zone (TAZ) data, origin-destination (OD) trip distribution, link volumes (AM, midday and PM peak), signal timing data, and transit ridership data. The network geometry data were obtained through aerial and street view maps and used for coding the network. The TAZ data, along with OD trip distribution and link volumes, were obtained from the Wasatch Front Regional Council (WFRC), and these data exist for the years 00 and 00 (forecasted). Signal timing data for signalized intersections are downloaded using Utah Department of Transportation i software, which allows a direct communication link to the field traffic controllers and control program databases. Traffic signals are coded simultaneously in VISUM and VISSIM. Transit ridership data, that also includes boarding and alighting information for transit stops within the network, are obtained from Utah Transit Authority (UTA). OD Matrix Correction and Model Calibration Before any detailed modeling was performed on the microscopic level, the macroscopic traffic assignment models in VISUM were calibrated using the actual link volumes for the year 00, for AM and PM peak periods. The traffic volume-based calibration required multiple iteration process until OD matrix was adjusted to fit the field data. The assignment analysis of the adjusted OD matrix from the Figure shows a satisfying correlation between link volume data from WFRC and the assigned volumes in VISUM. The calibrated OD matrix results in Figure are provided for the 00 PM peak period. FIGURE Traffic assignment calibration, PM Peak, 00.

8 I. Tasic, and M. Zlatkovic Performance Measures Calibrated matrix with static assignment of traffic volumes and assigned transit schedules and ridership information was exported from VISUM to VISSIM. The microsimulation environment allows for a very detailed evaluation of performance measures related to traffic efficiency: Level of Service (LOS) for intersections based on intersection delay Travel time and trip distance for representative trips on the major corridors Network performance through average speed, average number of stops, and total delay Since the imported network includes both auto and transit mode, VISSIM could measure average speed for both modes as an indicator of mobility. The additional performance measures that were observed to compare this base network with new network designs include the increase in trip redundancy and the number of cars rerouted from 00 West as the major corridor in the case study network. The local transportation agencies are interested in dispersing traffic through enhanced connectivity strategies, but also want to prevent high percentage of rerouted vehicles from the major arterials into the local streets of the case study network. The set of performance measures from microsimulation provides the information on the way enhanced connectivity impacts vehicular traffic, which can be further combined with the findings from the existing literature (e.g. relationship between connectivity and alternative modes of transportation); and utilized to evaluate the effects of connectivity strategies and select the optimal network designs that will accommodate transit, non-motorized and vehicular traffic. Street Connectivity Scenarios The advantage of the case study network shown in Figure is that it is in fact a grid-like network however the spacing between the streets does not encourage alternative modes (-). The case study network is bordered by arterials, where 00 W and 00 S are the major corridors, with a Bus Rapid Transit (BRT) line in place and plans for a major multimodal corridor construction in close proximity to the network. There are a lot of cul-de-sacs and disconnected links on the local streets inside the analyzed network. The combination of the grid-like structure with large spacing between major streets and disconnected links inside the network simplifies the task of modeling various connectivity levels on the network, as it was possible to model new connections in continuation of the already existing links and cul-de-sacs. The reviewed literature shows that denser street networks decrease the need for private vehicle use (-), but this study does not consider any mode shifts in order to account for the case of highest travel demand in the developed scenarios, which is expected right after the implementation of enhanced connectivity strategies. The design principles are focused on street spacing and traffic speeds on the existing and newly added corridors. Based on the recommendations from the literature, the intersection spacing goes as low as 00 feet (-). The goal of the proposed network modifications/connectivity improvements was not to eliminate driving or enforce other modes of transportation, but to enable alternative modes of transport and to make them a relevant part of the transportation options choice, especially for those people who cannot or choose not to drive. A total of ten new network design scenarios were modeled and evaluated with different connectivity levels and corresponding street widening alternatives. Figure shows five enhanced connectivity scenarios b to b, versus five street widening scenarios a to a. The length of new connections in scenarios b to b is equivalent to the length of added lanes in scenarios a to a. Five pairs of connectivity-widening alternatives are obtained in this manner.

9 I. Tasic, and M. Zlatkovic 0 0 The scenarios in Figure were designed for the comparison of impacts of different levels of network connectivity on traffic operations. Although the network is bordered by major arterials where posted speed limit ranges from - mph, the newly-added connections within the case study network were modeled to not exceed mph, following recommendations from the literature for local street networks (-). Each scenario and approach rendered a different traffic assignment in VISUM, and thus different vehicle inputs and routing decisions in VISSIM models. Traffic Calming Scenarios For the case of the highest proposed connectivity level (scenario b ), an additional scenario was developed to mitigate the effects of major inflow of traffic into the local street network, using reduced speed areas. The areas of reduced speeds inside the local network were modeled based on the recommendations from the traffic calming measures practice. Traffic calming studies are usually based on the empirical evidence, and analyzed for their safety effects. While previous studies found that traffic calming has positive effects on safety, their operational effects are rarely tested (). This is due to the fact that traffic calming is installed in neighborhoods, to lower the speeds of traffic. It is however important to examine the effects of these measures on the network wide level, especially in TOD environments. The equation from the U.S. Traffic Calming Manual was used to calculate the optimal spacing of traffic calming measures in the case study network, depending on the midpoint speed, street speed and low point speed (). The th midpoint speed represents the speed that is up to mph over the posted speed limit. Street speed is posted speed limit, while low point speed is the target speed that should be achieved through traffic calming installation. 0 Before using these calculations to allocate traffic calming effects in the form of decreased link speeds, the posted speed limits were compared with assigned traffic speeds on the existing network and network with increased connectivity. This comparison was used to identify the potential network areas where speeding might occur as the network density increases. With the described methodology, a total of scenarios were modeled and evaluated, including the existing base case scenario, five street widening scenarios, five street connectivity scenarios, and additional scenario with traffic calming effects. All developed scenarios were modeled for the existing data for 00 AM peak period ( AM to AM) and 00 PM peak periods ( PM to PM); as well as the forecasted travel demand data for 00 AM and PM peak periods. The results are presented and discussed in the following section.

10 I. Tasic, and M. Zlatkovic 0 a) Street widening scenarios Scenario a Scenario a Scenario a Scenario a Scenario a b) Street connectivity scenarios Scenario b Scenario b Scenario b Scenario b Scenario b Proposed street widening (Scenarios a) or new connections (Scenarios b) FIGURE Simplified street widening (a-a) and street connectivity scenarios (b-b).

11 I. Tasic, and M. Zlatkovic RESULTS AND DISCUSSION This section discusses the implications of connectivity on traffic operations on the part of the West Valley City network in Utah that represents a potential TOD spot according to Wasatch Front regional long-term transportation plan. The results are presented at the intersection, corridor, and network wide level. Intersection Performance Table presents the results of intersection LOS, comparing street widening and street connectivity scenarios for all periods of analysis. The intersection analysis shows that increased street connectivity does not impact intersection performance, as the critical intersections along the major transit corridors of 00 W and 00 S retain desired LOS for 00 periods. The LOS results are similar for street widening and street connectivity scenarios for 00 and 00 AM and PM peak periods, showing that street connectivity strategies could be an alternative for traditional capacity expansion strategies that usually involve street widening. Various levels of street connectivity have different effects on intersection delay for PM peak periods (Figure ). As street connectivity increases intersection delays also increase up to a certain point, but then start to decrease, when compared to base case scenario. This indicates that simple increase of street connectivity does not indefinitely benefit traffic progression, implying that network connectivity levels need to be optimized to achieve improved performance. The results provided in Figure, for the 00 PM and 00 PM peak periods show that addressing intersection designs, especially along the major arterials, could significantly improve the network performance. As the results at the intersection level indicate that street widening is not necessarily more beneficial than the enhanced connectivity, some alternative intersection design might be the solution for the increase in intersection delay for the forecasted travel demand for 00. Corridor Performance Travel times, speeds and LOS on the corridor level for the existing network, street widening and connectivity scenarios are analyzed for all segments between the major intersections in the case study network. The microsimulation models showed that the additional connections between the 00 W corridor and 00 West street do not cause the traffic to detour from this corridor and use other streets as alternatives in Southbound direction. The decrease in LOS and speeds, and increase in travel time along this corridor, with even only one additional street connection to the parallel arterial shows that more drivers would choose this corridor if more connections were provided. As shown in Figure, additional street connections to 00 W corridor decrease travelers delay in Northbound direction, while street widening scenarios result in lower delay in Southbound direction along this corridor. This implies that simple street widening or adding connections that feed into this corridor will not improve its performance, confirming the previous conclusions that intersection designs could resolve network performance issues on this location. Analogue to travel time results, average traveling speeds along 00 W increase in Northbound and decrease in Southbound direction, with the addition of new connections in the network. The 00 W will remain the major corridor in this area regardless of the network changes, however just like the intersection analysis, the results on the corridor level show that enhanced connectivity is a promising alternative to street widening approach.

12 I. Tasic, and M. Zlatkovic Scenario Scenario Scenario Scenario Scenario Scenario 00 AM 00 PM 00 AM 00 PM 00 AM 00 PM 00 AM 00 PM 00 AM 00 PM 00 AM 00 PM 00 AM 00 PM 00 AM 00 PM 00 AM 00 PM 00 AM 00 PM Intersection 00 W 00 S TABLE Intersection LOS Analysis 00 W 00 S 00 W 00 S 00 W 00 S 00 W 00 S 00 W 00 S 00 W 00 S 00 W 00 S a B C C A A C D B b C C B B A B C B a C C B A A B C B b C B B A A B D B a C C E A A C C B b C C E B A B C B a E C F A B B D F b F C F B A B F F a C C C A A C D B b C C B B A B C B a C C C A A B D B b C B B A A B D B a C C E A A C C B b C C F B B B C B a F C F A B B D F b F C F C B B F F a C C C A A C D B b C B B B B B C B a C C C A A B C B b C B B A A B C B a C C E A A C C B b C C E B B B C B a F C F A B B D F b F C F C B B F F a C C C A A C D B b C B B B B B C B a C C B A A B C B b C B B A A B C B a C C E A A C C B b C C E B B B C B a F C F A B B D E b F C F C B B F F a C C C A A C C B b C B B B A B C B a C C B A A B C B b C B B A A B C B a C C E A A B B B b C C E B B B C C a F C F A B B C E b F C F B B B F F

13 Intersection Delay (s) Intersection Delay (s) I. Tasic, and M. Zlatkovic 0.0 Intersection Delay Comparison for 00 PM Peak Period W 00 S 00 W 00 S 00 W 00 S 00 W 00 S 00 W 00 S Major Intersections 00 W 00 S 00 W 00 S 00 W 00 S Intersection Delay Comparison for 00 PM Peak Period Base Case Scenario b Scenario b Scenario b Scenario b Scenario b Base Case Scenario b Scenario b Scenario b Scenario b Scenario b W 00 S 00 W 00 S 00 W 00 S 00 W 00 S 00 W 00 S Major Intersections 00 W 00 S 00 W 00 S 00 W 00 S FIGURE Comparisons of intersection delays for base case, and increased connectivity scenarios for PM peak periods.

14 I. Tasic, and M. Zlatkovic a) Northbound direction 0 b) Southbound direction FIGURE Average travel time and speed along the 00 West Corridor (00 PM) Network Performance Traffic analysis of various street connectivity scenarios on the network wide level is given in Table and Figure. Enhanced street connectivity increases the overall network delay, when compared to street widening and base case scenarios. The complete network analysis shows that networks with enhanced connectivity accommodate more vehicles during the same period of time. So it is a trade-off between capacity and total network delay, whether the existing state of the network will be kept or connectivity will be increased for the current traffic conditions. Considering the travel forecasts for 00 AM and PM peak periods however, enhanced connectivity contributes with up to 0 seconds to average delay per vehicle, while it accommodates about 000 vehicles more than the base case or street widening scenarios. So for

15 I. Tasic, and M. Zlatkovic 0 0 the future network modifications, street connectivity with alternative intersection designs might be the network development that could address the demand. Table also shows network analysis of scenarios that include traffic calming with the highest level of street connectivity applied (Scenario b ). Traffic calming measures modeled in this way affect 00 PM peak period the most, with the highest delay values. Further research needs to be done with various combinations of street connectivity and traffic calming implementation to determine the optimal network density and speeds. Our results show that traffic calming has influence on the entire network, even though it is only applied to local streets. TOD does not necessarily require traffic calming, but in the case where the intersection density alone does not decrease traffic speeds to encourage alternate modes, calming traffic is both and efficient and non-expensive way of preventing high speeds in the environment that should be pedestrian-friendly. Figure shows that the total distance traveled is generally higher in the PM peak period for both 00 and 00 modeled scenarios. In addition, the total distance traveled is slightly higher in street connectivity than in street widening scenarios, but this difference is not significant, implying that enhanced street connectivity does not increase the vehicle miles traveled significantly. The network-wide results show that enhanced connectivity opens new routes and provides better dispersion of intra-zonal traffic, without rerouting external-external trips from the major arterial. As connectivity increases, network designs with enhanced connectivity accommodate more traffic than designs with street widening. However, none of the proposed solutions will meet the 00 traffic demand unless mode shift occurs, implying the need for transit enhancements in the area. TABLE Network-Wide Performance, Street Widening vs. Enhanced Connectivity 00 AM Base a b a b a b a b a b Traffic Calming Total number of vehicles,00,0,,0,0,0,,0,,00,, Average delay time per vehicle (s) 0 Average number of stops per vehicles Total delay time (h) Average speed (mph) Total travel time (h),0,,,,,,,,,,, Total distance traveled (mi),0,0,,00,0 0,, 0,,,0,, 00 PM Base a b a b a b a b a b Traffic Calming Total number of vehicles,,,,,0,,,,,,, Average delay time per vehicle (s) Average number of stops per vehicles Total delay time (h) 0 0 Average speed (mph) Total travel time (h),,0,,,0,,,,0,,, Total distance traveled (mi),00,,, 0,,0,0,0,,0 0, 0, 00 AM Base a b a b a b a b a b Traffic Calming Total number of vehicles,,,, 0,,,00,0 0,0,,, Average delay time per vehicle (s) 0 0 Average number of stops per vehicles Total delay time (h), Average speed (mph) Total travel time (h),0,0,,0,,0,0,0,,0, 0, Total distance traveled (mi),0,,,0,0,0,0,0,,,, 00 PM Base a b a b a b a b a b Traffic Calming Total number of vehicles,0,,,0,,,,0,,,,0 Average delay time per vehicle (s) 0 0 Average number of stops per vehicles Total delay time (h),,,0,,0,0,,,,,, Average speed (mph) Total travel time (h),,0,,,,,,,0,,00,00 Total distance traveled (mi) 0,,,, 0,,0, 0,0, 0,,,

16 Total Distance Traveled (miles) Total Network Delay (h) Total Distance Traveled (miles) Total Network Delay (h) I. Tasic, and M. Zlatkovic 0,000 0, Scenarios , , , , , Base a b a b a b a b a b Scenarios 0 Total Distance Traveled AM Total Network Delay AM Total Distance Traveled PM Total Network Delay PM 00,000 0, Scenarios,000,000 0,000,000 0,000,000 0,000,000 0 Base a b a b a b a b a b Scenarios 0 Total Distance Traveled AM Total Distance Traveled PM Total Network Delay AM Total Network Delay PM FIGURE Total distance traveled and network delay for base case, connectivity, and street widening scenarios for 00 and 00, AM and PM peak periods

17 I. Tasic, and M. Zlatkovic CONCLUSIONS AND RECOMMENDATIONS Highly connected street networks increase accessibility for multimodal transport, but their effects on the efficiency of still dominant vehicular traffic are rarely addressed. This paper examines the effects of different strategies related to street network patterns on traffic operations of a test network in West Valley City, Utah. The analysis of our Base Case Scenarios shows that PM peak period will be more critical in 00, especially for 00 W & 00 S and 00 W & 00 S intersections. Both average per vehicle and total delay on the network wide level increase by more than 0% in AM and 00% in PM peak period, when we compare 00 and 00, which means that no build solution is not an option, as expected. Comparison of travel times and speeds on different segments for 00 and 00 shows significant increase in travel time for only one of segments we compared on our test network, meaning that new network designs for 00 need to focus on intersection operations. Increased street connectivity without improving intersection operations will not accommodate traffic demand for 00 PM peak period, under the assumption that mode shift does not occur. Comparing street connectivity scenarios for different network segments between main intersections, street widening and enhanced connectivity show similar results, implying that enhanced connectivity could be a good alternative approach for the corridors. Network designs with higher levels of street connectivity show similar or better performance on the intersection, corridor, and network-wide level, than designs with street widening. Increased connectivity, as an alternative to street widening slightly increases total distance traveled, but the delay values on the network wide level show that designing the network with multiple connections, rather than simply widening the arterials, would be a good alternative. Adding traffic calming measures to the network design with increased connectivity, increases total network delay. All these conclusions should be observed with the consideration that enhanced network designs are evaluated under the assumption that changes in mode split as a consequence of increased connectivity did not occur. It was important to evaluate the developed scenarios under this assumption, as the common issue that typical suburban areas might face while attempting to develop in multimodal environments is dealing with increased congestion right after street network modifications take place, while desired mode shift is yet unachieved. This paper adds to the existing literature that examines the relationships between the level of street network connectivity and traffic operations, and shows that enhanced street connectivity has the potential to compete with traditional capacity expansion strategies such as street widening. The future research efforts should address the impacts of various network designs modeled here, with the inclusion of different mode split scenarios. ACKNOWLEDGEMENTS The authors would like to thank the Utah Transit Authority for funding this study in its first phase; Mountains Plains Consortium for supporting the continuation of this study; the Avenue Consultants, Dr. Reid Ewing, Laney Jones (Lochner), Dr. Muhammad Farhan and Andy Li for providing their insights and suggestions for the methodology developed in this study; Wasatch Front Regional Council, Resource System Group, and Utah Department of Transportation for providing the needed data for this study.

18 I. Tasic, and M. Zlatkovic REFERENCES. TDM Encyclopedia. Roadway Connectivity: Creating More Connected Roadway and Pathway Networks. Victoria Transport Policy Institute, Updated March, 0. Wasatch Choices 00: A Four County Land-Use and Transportation Vision. Envision Utah. Accessed online: in January, 0.. Initiative for Walkable Salt Lake. Accessed online: in January, 0.. SixtyNine Seventy, The Spaces Between: An Urban Ideas Competition. Accessed online: in February, 0.. Salt Lake City Downtown Master Plan. Initiative for Walkable Salt Lake. Accessed online: Conditions-Analysis-Ch--0.pdf in April, 0.. Salt Lake City Mid-Block Walkways Design Guidelines. Initiative for Walkable Salt Lake. Accessed online: Mid-block-Walkways-Design-Guidelines.pdf in April, 0.. Handy, S., R. G. Paterson, K. Butler. Planning for Street Connectivity: Getting from Here to There. Planning Advisory Service Report No., American Planning Association, 00.. Charter of the New Urbanism. Congress for the New Urbanism. Ratified.. Dill. J. Measuring Network Connectivity for Bicycling and Walking. Presented at rd Annual Meeting of the Transportation Research Board, Washington, D. C., Tresidder, M. Using GIS to Measure Connectivity: An Exploration of Issues. Portland State University, School of Urban Studies and Planning, December, 00.. Yi, C. Utilizing GIS to Measure Street Connectivity and Pedestrian Accessibility: Comparing Cul-De-Sac and Grid Neighborhoods in the Houston Metropolitan Area. GIS-T Student Paper Contest, 00.. Tal, Gil, and Susan Handy. "Measuring Nonmotorized Accessibility and Connectivity in a Robust Pedestrian Network." Transportation Research Record: Journal of the Transportation Research Board, no. - (/0/ 0): -.. Ewing, R. (). Best Development Practices; Doing the Right Thing and Making Money at the Same Time. Planners Press.. Ewing, R., and R. Cervero. Travel and the Built Environment: A Meta-Analysis. In Journal of the American Planning Association, Vol., Issue, June 00, pp.

19 I. Tasic, and M. Zlatkovic Designing Walkable Urban Thoroughfares: A Context Sensitive Approach. Institute of Transportation Engineers and Congress for the New Urbanism. Publication No. RP-0A. Washington, D.C., 00.. Smart Transportation Guidebook: Planning and Designing Highways and Streets that Support Sustainable and Livable Communities. New Jersey Department of Transportation and Pennsylvania Department of Transportation. March, 00.. Ewing, R., and King, M. Flexible Design of New Jersey s Main Streets. Rutgers University and New Jersey Department of Transportation. Trenton, New Jersey... Handy, S.L. and D.A. Niemeier. Measuring Accessibility: An Exploration of Issues and Alternatives. In Environment and Planning A No.,, pp. -.. Scoppa, M., S. French, J. Peponis. The Effects of Street Connectivity upon the Distribution of Local Vehicular Traffic in Metropolitan Atlanta. Proceedings of the th International Space Syntax Symposium, Stockholm, June, Garrick, N.W., and W.E. Marshall. Street Network Types and Road Safety: A Study of California Cities. In Urban Design International, 00, page -.. Ewing and Dumbaugh, The Built Environment and Traffic Safety: A Review of Empirical Evidence, In Journal of Planning Literature, 00.. Handy, S. Urban Form and Pedestrian Choices: Study of Austin Neighborhoods. In Transportation Research Record: Journal of the Transportation Research Board, No., Transportation Research Board of the National Academies, Washington, D.C.,, pp. -.. Gallagher, P., W. Marshall, and C. Atkinson-Palombo. Missing Links. In Transportation Research Record: Journal of the Transportation Research Board No., Transportation Research Board of the National Academies, Washington, D.C., 0, pp. -.. Dill, J., and K. Voros. Factors Affecting Bicycling Demand: Initial Survey Findings from the Portland, Oregon, Region. In Transportation Research Record: Journal of the Transportation Research Board, No. 0, Transportation Research Board of the National Academies, Washington, D.C., 00, pp. -.. Foletta, N., Nick V., and B. Grandy. Factors That Influence Urban Streetcar Ridership in the United States. In Transportation Research Record: Journal of the Transportation Research Board, No., Transportation Research Board of the National Academies, Washington, D.C., 0, pp. -.. Rajamani, J., C. Bhat, S. Handy, G. Knaap, and Y. Song. Assessing Impact of Urban Form Measures on Nonwork Trip Mode Choice after Controlling for Demographic and Level-of-

20 I. Tasic, and M. Zlatkovic Service Effects. In Transportation Research Record: Journal of the Transportation Research Board, No., Transportation Research Board of the National Academies, Washington, D.C., 00, pp. -.. Cairns, S., Disappearing Traffic? The Story So Far. In Proceedings of the Institution of Civil Engineers, Vol., Issue, The Institution of Civil Engineers, U.K. March, 00, pp. -.. Evaluating Non Motorized Transportation, Victoria Transportation Policy Institute, January, 0.. Alba, C. B. and E. Beimborn. Analysis of the Effects of Local Street Connectivity on Arterial Traffic. Presented at th Annual Meeting of the Transportation Research Board, Washington, D. C., Ortigosa, J., M. Menendez, and A. K. Bodenbender. Link removal on a Grid Street network. th Swiss Transport Research Conference. May, 0.. Huang, H.F., and M.J. Cynecki. The Effects of Traffic Calming Measures on Pedestrian and Motorist Behavior. Report No. FHWA-RD Federal Highway Administration. August, 00.. Ewing, R., and S. Brown. U.S. Traffic Calming Manual. American Planning Association. 00.