(2) Impact of Parking Maneuvers on Space Mean Speed And Average Travel Time

Full text

(1)IMPACT OF PARKING MANEUVERS ON SPACE MEAN SPEED AND AVERAGE TRAVEL TIME Ermias Tesfaye Adula KTH, May 2011.

(2) Impact of Parking Maneuvers on Space Mean Speed And Average Travel Time. 2.

(3) Impact of Parking Maneuvers on Space Mean Speed And Average Travel Time. Abstract Parking maneuver is one among the numerous factors which affect the traffic movement. In this thesis work the impact of parking maneuvers and double parking on the traffic performance of urban street segments is studied. In order to analyze the traffic performance, the impact on the travel time and space mean speed is studied. Two approaches are used for the study; the Micro analysis and Macro analysis. In the Micro analysis the study is not time bounded. The impact of individual vehicles is studied against the event of interest. These events include inbound parking maneuvers, outbound parking maneuvers, double parking and the load/unload activities. On the other hand for the Macro study the events in a five minute interval are studied for eight and a half hour for each site. Each five minute interval is taken as a single data input for the regression analysis. To study the impact on travel time speed trajectory is used and to study the space mean speed regression analysis is used. According to this study the parking maneuvers give rise to a speed reduction of almost 2 Km/hr and double parking 7km/hr. Similarly, the load/unload maneuvers cause a speed reduction of 3Km/hr. Normally two basic tasks are included in this thesis work; data reduction and data analysis. The raw data for this thesis work is a video recording of 8 and half hours for four different spots. Regression analysis and speed trajectory are the main analysis approaches employed for the data analysis. Speed trajectory is used for the Micro analysis and regression analysis is used for the Macro analysis.. 3.


(5) Impact of Parking Maneuvers on Space Mean Speed And Average Travel Time. Acknowledgments This thesis project is part of the Parking project which is carried out at the traffic and logistics department of KTH. This Parking study, which is focusing on the selected streets in the central areas of Stockholm, is carried out by a team of professionals lead by Professor Karl-lennart Bång. With all his grace thanks to the Almighty God I finally made it through with the unexplainable collaboration and assistance of numerous people. First, I would like to thank and appreciate Professor Karl-lennart Bång for his best guidance and wisdom sharing. As my supervisor and chairman of the project, his guidance put on the right alignment to accomplish my work. Secondly, I would like to thank the project team members Ary-Pezo Silvano (Phd. Candidate), Dr. Albania Nissan and Dr. Azhar Al-Mudhaffar for their continuous assessment. They were so perfect in following every move I did. My special thanks go to Professor Harris Koutsopoulos for his time and best ideas in solving my major problems. Finally, I would like to thank my wife Tsedey Yeshitila Belayneh for all the caring and support. Lastly, I would like to mention my family back home. This paper work is dedicated to my late Dad, whose dream drove me extra miles in my academic life.. 5.


(7) Impact of Parking Maneuvers on Space Mean Speed And Average Travel Time. Table of Contents 1. Introduction .................................................................................................................................... 9 1.1. Background ............................................................................................................................. 9. 1.2. Parking Definition.................................................................................................................... 9. 1.3. Objective ............................................................................................................................... 10. 1.4. Scope and Limitation............................................................................................................. 10. 2. Literature Review .......................................................................................................................... 11. 3. Methodology................................................................................................................................. 17 The data used for this project is a video recording which lasts for about eight and half hour for each site. This data was collected in September 2009 by the division. From this raw data intensive observation was made to collect all the necessary information. ..................................................... 17 3.1. 3.1.1. Free Flow Speed ............................................................................................................ 17. 3.1.2. Parking Maneuver ......................................................................................................... 17. 3.2. 4. Micro Study ........................................................................................................................... 17. The Macro Study ................................................................................................................... 19. 3.2.1. Data Analysis/Processing .............................................................................................. 20. 3.2.2. Regression Analysis ....................................................................................................... 21. Description of Case Study (Data Analysis) .................................................................................... 27 4.1 Case Study One ........................................................................................................................... 27. 5. 4.2. Case study Two ..................................................................................................................... 32. 4.3. Case study Three ................................................................................................................... 34. Results and Discussion .................................................................................................................. 36 5.1. Micro Analysis ....................................................................................................................... 36. In the Micro analysis the data reduction is not done in a time bound manner. Some of the interesting events of the same kind are categorized together to see their impact. The interference of some other disturbing factors than the events of interest was more challenging to exclude...................................................................................................................................... 36 5.1.1. Case Study One ............................................................................................................. 36. 5.1.2. Case Study Two ............................................................................................................. 46. 5.1.3. Case Study Three........................................................................................................... 54. 5.2. Macro Analysis ...................................................................................................................... 62. 5.2.1. Evaluation of the Regression Analysis .......................................................................... 62 7.

(8) Impact of Parking Maneuvers on Space Mean Speed And Average Travel Time 5.2.2. Evaluation of the Initial Hypotheses ............................................................................. 63. 6. Discussion...................................................................................................................................... 66. 7. Conclusions ................................................................................................................................... 67. 8. Further Study ................................................................................................................................ 68. 9. Bibliography .................................................................................................................................. 69. 8.

(9) Impact of Parking Maneuvers on Space Mean Speed And Average Travel Time. 1 Introduction 1.1 Background With the fast growth of automobile use in the early 19th century decision makers and urban planners visualize the problems associated with the availability of proper parking space. For this reason it has become one of the major parts of a modern urban set up. In previous times parking related problems appear to be seen in the city centers and in areas with higher public movements. But, nowadays it is a concern throughout the urban area, even with those having moderate public activities. Parking activity gives the cities and suburbs the picture that they hold; it has impact on the traffic congestion, traffic operation and also determines the reliability of transit systems. The presence of parking space also puts a great deal of influence on the mode choice and route selection of travelers. On-street parking in situations without proper parking space could increase the accident risks and also could obstruct emergency vehicles. Additionally, in the mid 1930’s it was realized that it will be a major source of revenue (Childs, 1999). Many attempts were done to implement and propose different measures in order to create a better living environment and make people’s interaction so smooth. A well studied and designed parking policy results in an efficient transport system, lower emissions and inclusive urban layout (Marsden, 2006). Besides, accessibility and parking suitability are among the most decisive factors that affect shoppers’ destination choices (Innes D.et al, 1990). Parking policy is a vital issue for economy, sustainability and suitability of cities to live in. In Stockholm different policies have been implemented to improve the traffic condition in the inner city. One of the activities which affect the traffic condition in the innercity is the parking activity. For this purpose this study is carried out to evaluate the impact of the parking activities in the traffic performance of the streets in Stockholm.. 1.2 Parking Definition Usually vehicles have an origin and destination. Typically private cars usually have their origin in residential areas and destination in central city areas. Those cars have to be parked in the city areas which are flocked with cars. One point that we always have to bear in mind is that cars are parked for most of their life time. Various types of designs and facilities have been introduced to park cars. Literally, parking places can be categorized as: - Off-street parking indoors - Off-street parking outdoors - On-street parking (with fee or no fee) Off-street parking indoors 9.

(10) Impact of Parking Maneuvers on Space Mean Speed And Average Travel Time. This is a sort of facility which is designed to store cars out of the road network. They can be in the form of a basement which is built underground, level ground or a multi-story building is allocated for it. Mostly they are located in central business areas where there is a high demand for parking and less available space. Off-street parking outdoors These are parking lots at the ground level. They are quite common in office areas, residential areas, and big shopping malls or supermarkets. Despite the fact that they appear at the ground level they do not have a direct interaction with the traffic stream. They have typical layout design which can accommodate as many vehicles as possible. They also have enough drive way and medians. On-street parking outdoors This in most literatures is called curb side parking. It is a parking on the roadside bordering the traffic stream. Mostly these types of parking activities happen to be free of charge in residential areas. On the other hand in central areas there are parking meters for the drivers to pay parking fees. On-street parking outdoors is the type of parking that will be studied in this thesis work.. 1.3 Objective The aim of this study is to analyze the impact of parking maneuvers and double parked vehicles on the traffic performance of urban roads. Specifically, this thesis work studies the impact of those mentioned events on the speed and travel time of an urban street segment in Stockholm. For this purpose two approaches will be employed; Macro analysis and Micro analysis. In the Micro analysis the impact of individual maneuvers is studied whereas in the Macro analysis the impact of the events is studied in an aggregated approach. In doing so, the interference of other types of maneuvers or some other events will be excluded.. 1.4 Scope and Limitation At the start of this project the impact of bicyclists and pedestrians was studied. But in the preliminary assessment these parameters were not found significant. The bicyclists use their own lane unless obstructed and the number of pedestrians crossing not on the zebra cross is very few. For those reasons the two variables are excluded from the study. The independent variables used in the Macro analysis are flow, maneuvers, double parking, load/unload and number of lanes. These variables are found significant. The other limitation is the impact of aggregation and averaging. Due to some constraints the time segment could not be narrower than five minutes in the Macro study the analysis. When the events in a five minute interval are average, some events happen to be covered up and their impact will not be seen more significantly than those without any event. Usually when events occur like with the red traffic light, bunching of vehicles happen. In this study the bunching vehicles are not analyzed. 10.

(11) Impact of Parking Maneuvers on Space Mean Speed And Average Travel Time. 2 Literature Review There are lots of literatures about parking issues which are related to parking behavior, parking policies, economic analysis of garages, optimization of garages, etc. However, there are very few literatures regarding the influences of on-street parking on traffic performance, traffic safety, logistic (Load-Unload), accessibility and the environment. The main objective of this literature review is to revise the papers on how on-street parking activities influence the speed of adjacent traffic stream and capacity of the road where they are situated. Normally there is a direct relation between speed and capacity in the intercity roads; as the speed reduces so does the capacity. It has been advocated that on-street parking affects the capacity and speed of the streets. An intensive search on different sources on this subject matter is summarized in the following. In literatures it is stated that on-street parking reduces the capacity for various reasons. To mention some; the on-street parking lane could have been dedicated to the through traffic and it also reduces the capacity as it reduces the speed of the vehicles due to the side friction. This is also called roadside characteristic and is demonstrated by the models elaborated below. Bång,_K._L. (1995) developed speed-flow relationships and road capacity () in the Indonesian Highway Capacity Manual. On-street parking was taken in to account in the estimation process. Capacity was related to explanatory variables such as road function, roadside friction and road width. For the roadside friction in the urban streets, on-street parking was considered with the number of parking maneuvers occurring in the road segment (//). In order to estimate the parameters of the models linear regression was applied. The capacity model is:

(12) . 1.1. Where . = Actual capacity. . = Ideal capacity.

(13). = Road width adjustment factor. . = Directional adjustment factor. . = Road friction adjustment factor. Different road characteristics were considered in the roadside adjustment factor( ); these include pedestrians(Ped), on-street parking maneuvers(PSV), number of bus stops and slow moving vehicles (SMV) with different weights. The following equation gives an 11.

(14) Impact of Parking Maneuvers on Space Mean Speed And Average Travel Time. empirical relationship which tells the dependence of road frictional adjustment factor on the factors mentioned above: FCsf=Ped*0.6+PSV*0.8+SMV*0.4 According to Bång (1995), the roadside activities significantly affect the speed-flow relationship of the road links. This will result in the speed and capacity reductions. In the graph in Figure 2.1 below it can be seen how speed and capacity are reduced due to side friction events.. Figure 2.1: The relationship of speed and capacity with side friction, Source: Bång, K.L (1995) Another approach was made by Highway Capacity Manual (HCM 2010-Chapter 17). For urban street segments it considers geometry, traffic and control characteristics to compute urban street segments performance and Level of Service (LOS). Four modes are considered in the methodology: pedestrian, automobile, bicyclist and transit. In the computation of LOS for pedestrian and bicyclist modes on-street parking is considered directly; yet the HCM 2010 does not consider on-street parking for automobile mode. It is left to the analyst to complement the analysis adding typical methodology if available to compute it’s influence on the traffic performance. 12.

(15) Impact of Parking Maneuvers on Space Mean Speed And Average Travel Time. Ibeas-Portilla, A._et_al (2009) evaluated traffic performance due to on-street parking with a different approach. To quantify the influence of parking maneuvers and bad parked cars on average link journey times as a function of the duration of the events and flow the authors applied M/M/∞ queuing model. To calibrate the model data was collected from the city Santander in Spain. Validation was done with micro simulation (AIMSUM). Parking maneuvers and badly parked cars are introduced as events where the lanes do not work or work at a lower normal rate of service. Parking maneuvers are introduced as high frequency short duration event (10, 20 and 30 maneuvers per hour) and badly parked cars are introduced as time events (15 min, 30 min, 45 min and 60 min). The results show an increase in average journey time up to by 15, 24 and 39% for 10, 20 and 30 maneuvers respectively; and a reduction in capacity of 6, 10 and 16% for maneuvers of 10, 20 and 30 respectively. Similarly the journey times increase by 57 and 107% for badly parked vehicles for 15 and 30 min respectively; and capacity reduce by 13 and 27% respectively. When the event lasted for 60 min the capacity reduction reached 55%. Studies illustrate that speed reduces with the presence and amount of on-street parking for urban streets (Shoup 2006, Childs 1999). All of the studies carried out linear regression analysis to find the average speed on the streets. They took in to account road characteristics as predictors of the dependant variable (Space mean speed). In all the studies the presence of on-street parking was found to be an important factor to reduce the speed of the motor vehicles. Aronsson(2006) carried out a study to estimate the mean speed on different types of streets. She applied linear regression model for the analysis. Different types of streets are categorized according to the service they provide; Arterial, Suburban and Urban streets. Various street characteristics such as traffic flow, on-street parking, crossing pedestrians, and number of lanes are tested in a linear regression model to estimate their impacts on the space mean speed. The result showed that on-street parking has negative impact on the space mean speed by (-5.54). In a similar study, Nordström (2006) took into account mixed traffic in the city centers and applied linear regression model to estimate the space mean speed as dependant variable. According to the study, the presence of on-street parking was found to reduce the average speed by 3km/h.. 13.

(16) Impact of Parking Maneuvers on Space Mean Speed And Average Travel Time Mean Speed-Urban Streets Predictors Estimates Intercept 39.8 Flow Pedestrian. -0.202 -0.237. Lanes. 5.24. bic_sep. 4.73. Parking. -5.54. Table 2.1: Mean speed model developed at KTH by Aronsson (2006), coefficients of the regression analysis.. Mean Speed-Urban Streets Predictors Estimates Intercept 45.4 Flow bic_sep Parking. -0.171 -6.31 -2.93. Table 2.2: Mean speed model developed at KTH by Nordström (2006), coefficients of the regression analysis. A study about speed-flow relationship and roadside characteristic was also carried out by Aronsson (2006). And it was reported that Wang et al (2006) built a model with which the operating speed in urban streets is estimated. The author used different roadside characteristics in the regression analysis to estimate the parameters from second-by-second in-vehicle GPS data from two hundred randomly selected vehicles in Atlanta, Georgia, USA. The result shows that on-street parking reduces the speed by 3.19 miles/h. See the Table 2.3 below for the estimates. 14.

(17) Impact of Parking Maneuvers on Space Mean Speed And Average Travel Time Mean Speed-Urban Streets Intercept. 31.56. Number of lanes. 6.46. Number of roadside objects/mile. -0.1. Number of intersection/mile. -0.08. Number of driveways/mile. -0.05. Kerb indicator. 3.01. Pavement indicator. -4.26. Parking indicator. -3.19. Land use 1. 3.31. Land use 2. 3.27. Table 2.3: Operating speed model developed by Wang (miles/hour). With another approach Tivector(2009) studied the influence of on-street parking on the average speed of urban street. On the study the streets were evaluated for one-sided and two-sided on-street parking. The results of the study are consistent with the results of the studies above showing a higher speed reduction for two-sided on-street parking. For twosided parking the speed reduction is 8.9km/h while for one-sided it is 4.5km/h. From the results it is obvious that the impact is doubled on the two sided on-street parking when compared to the one-sided on-street parking. Table 2.4 below shows the speed model developed by Trivector; the effect of one-sided and two sided on-street parking estimates.. Mean Speed-Urban Streets (50 km/h) Predictors Estimates Intercept 38.7 Local 0 Street type Main 7.6 One-sided -4.5 Parking Two-sided -8.9. Table 2.4: Speed Model by Trivector (2009) (speed limit = 50 km/h) Hansen (2007) conducted a similar study in US to find out the effect of the roadside environment on the chosen speed. Data was collected from 272 road segments classified by the urban roads and highways. Linear regression was used to analyze the data; analysis of 15.

(18) Impact of Parking Maneuvers on Space Mean Speed And Average Travel Time. the variance was also employed to identify the significant variables. The variables used as independent variables were pavement width, shoulder width, lane width, sidewalk width, planting strips, building setbacks, access density and land use types. Table 2.5 shows the results for the urban model; the on-street parking has been divided in to two levels of utilization as heavy parking and light parking. The demarcation is 30% utilization. Whenever the block is occupied by parked cars for the space greater than 30% called heavy parking, otherwise light parking. They report that heavy parking has significant effect reducing the chosen speed by 2.33mph. Table 2.5 below shows the predictors and estimates of the model developed by Hansen (2007). Mean Speed-Urban Streets Predictors Estimates 29.94 Intercept Speed limit 35 mph. 2.53. Speed limit 30 mph. 2.09. Suburban residential. 7.32. Urban residential. 5.75. General urban. 4.83. Parking >30%. -2.33. Table 2.5: Speed Model by Hansen (2007) (miles/h). 16.

(19) Impact of Parking Maneuvers on Space Mean Speed And Average Travel Time. 3 Methodology The data used for this project is a video recording which lasts for about eight and half hour for each site. This data was collected in September 2009 by the division. From this raw data intensive observation was made to collect all the necessary information.. 3.1 Micro Study 3.1.1. Free Flow Speed. According to the literatures free flow speed is the term used to describe the mean speed of a vehicles in a segment when there is no congestion or adverse effect,(e.g. bad weather). For this study the free flow speed is the speed of vehicles without any obstruction of the parking maneuvers, pedestrians, bicyclists and other events which could disturb the smooth ride of vehicles. For this thesis purpose two case studies are carried out. As per the requirements for free flow speed condition vehicles were selected fairly from all the time periods. A minimum of 20 vehicles were selected for each case study. For the micro study the vehicles that are under the ideal free flow condition were considered. In the micro analysis a minimum of 20 vehicles were observed. The average travel time and space mean speed was calculated. The procedures followed are: 1. The time to travel every 10 meter segments was registered 2. The speed at every 10 meter points was calculated 3. The average travel time at each 10 meter segment was calculated; the average of all the observations at subsequent 10 meters is calculated for each category. 4. The average speed at each interval was calculated 5. Finally the (time-space) and speed profile graphs for each type of event were plotted using the free flow condition as a base line.. 3.1.2. Parking Maneuver. As it was already explained in the objective part, the main focus of this thesis is to see the impact of road side events e.g. parking maneuver, double parking and load/unload maneuver on the traffic performance of the urban street. As it is obvious from experience, with the occurrence of parking maneuvers the vehicles behind will be held back from their normal driving condition. Practically, it is not an easy task to isolate whether an impact has come typically from a parking maneuver or some other events in a flock of vehicles which are on a dynamic situation. It needs a careful observation for the whole segment. From the raw data, the video recording which lasts for an eight and half hour duration for each site, all the events were recorded. For the micro study out of the whole events recorded those which are affected only by the event of interest are categorized under a specific group. Then 17.

(20) Impact of Parking Maneuvers on Space Mean Speed And Average Travel Time. the average impact of the vehicles was calculated. As the study is to emphasize the impact of the parking maneuvers on the traffic performance of the street, the data was filtered in order to exclude the condition in which the traffic is affected by other factors like the traffic light when it is at the red. On the streets there are many events which disturb the flowing traffic. These disruptions make the drivers reduce their speed from their desired driving speed. Some of the conditions besides the parking maneuver under which the vehicles might also be disturbed: 1. Queue due to signal Some vehicles make a speed reduction due to the parking maneuvers. But after they overtake the vehicle which is making the maneuver they will then be held once again due to the red traffic light at the end of the block. In addition to this, in some cases the events which involve parking maneuver were also accompanied by a queue which was developed due to the traffic light. In these situations the drivers will be affected by multi events, and it will not be an easy exercise to distribute the effects to their particularly causing event. Thus, reduction in the traffic performance is not a pure impact of the parking maneuver. For this reason the vehicles in this condition were excluded from the data set. 2. Pedestrians’ interference On the zebra cross the pedestrians have the right of way to cross the street. On the contrary to this, some pedestrians cross the road at locations other than the zebra cross. Specially half way the block pedestrians mostly prefer to cross directly. Especially when the vehicles appear to be in a queue and there exists a reasonably higher head-way, it appears to be inviting. Similarly, in the course of this study few of the parking maneuvers have pedestrians crossing the road simultaneously. This condition definitely affects the drivers’ perception. For this reason, the maneuvers under this situation are excluded. 3. Bicyclists On all of the study sites selected there exists a bicycle lane. But in some cases the bicyclists jump in to the motorized vehicles lane. This condition is approached in two ways. When the parking lane is already occupied and vehicles will be forced to make a double parking and they always block the bicycle lane totally. This will force the bicycles to mix with the traffic and affect the vehicular movement. The root cause of this event is double parking. Therefore, these events are considered under double parking. But vehicles which are affected simply by the interference of bicycles without a double parking event are excluded.. 18.

(21) Impact of Parking Maneuvers on Space Mean Speed And Average Travel Time. 4. Isolated parking events This scenario includes the vehicles which are not disturbed by events other than parking maneuvers, double parking or load/unload condition maneuvers. If these events were not there the vehicles would have been on a normal driving condition. Those are the vehicles which are considered in this study. Out of the whole parking maneuvers happening throughout the day limited number of vehicles were found in this situation.. For the micro study, using the free flow line as a base, the deviation is measured for each category of events. These graphs are presented in Chapter 5, results and discussions. They demonstrate the delay of the vehicles. More discussion could be found on the result part. Sometimes, maneuvers happen while there is a long distance gap (distance head-way) so that the maneuvers will not have a significant effect on the smooth flow of the traffic. From the observation on the video it can easily be observed that drivers take enough time to have good distance headway when they are making out-bound maneuvers. This happens because they will have enough time to think about the proper gap acceptance which is not the case for inbound maneuvers. The presence of bicycle lane also helps for this phenomenon. It helps as a buffer area for the out bounding vehicles till they get a reasonable distance-head way. So the out bounding maneuvers appear to have a bit lesser effect on the traffic stream. In some cases the maneuvers with reasonably higher distance-headway are excluded as they have negligible impact on the following car. For the vehicles under Isolated parking maneuver condition the following procedures were carried out 1. The time to travel every 10 meter segments was registered 2. The speed at every 10 meter points was calculated 3. The average travel time at each 10 meter segment was calculated; the average of all the observations at subsequent 10 meters is calculated for each category. 4. The average speed at each interval was calculated 5. Finally the (time-space) and speed profile graphs for each type of event were plotted using the free flow condition as a base line.. 3.2 The Macro Study In this approach all the vehicles are included in the data analysis. In the events where there happens to be no parking maneuver, double parking or load/unload maneuver, the travel time was measured excluding the effect of other speed reducing factors like crossing pedestrians, bicyclists and other unusual events like the over speeding of emergency 19.

(22) Impact of Parking Maneuvers on Space Mean Speed And Average Travel Time. vehicles and police car chase. Sometimes due to the fact that the area is CBD (Central business district) some drivers happen to be biased to choose their free flow speed. Those vehicles were excluded from the analysis due to the fact that they will affect the final result. And for the macro study the travel time was averaged for the vehicles which are not affected by any event and the traffic light. For the macro analysis, the ultimate target is to analyze the situation in every five minute intervals and use each five minute interval as a single data for the regression analysis. 3.2.1. Data Analysis/Processing. For the Macro study all the maneuvers will be considered. In this case all the maneuvers will be aggregated to five minute time intervals. The observed time segment (eight and half) is sub-divided in to a five minute interval. Each site will have about 102 intervals. But sometimes there are intervals with a less suitable condition to be considered as a data. These types of data were excluded. The following general procedures were implemented on the data reduction: -. -. All the vehicles passing through the street segment were counted. With this the flow for each of the five minute intervals was calculated. The average travel time in each five minute time interval was calculated. The average speed was calculated using the length of the street segment. The maneuvers were summed up under each category; the inbound, reverse inbound and outbound maneuvers are taken under the category of maneuver, the presence of double parking and Load/Unload activity was checked under each time interval. In order to analyze the data, regression was applied. In the regression analysis the impact of different variables on speed was analyzed. In this model building the dependent and the independent variables need to be selected. Generally a table containing speed, flow, maneuvers, double parking, the load/unload, lane width, and number of lanes was developed.. For this analysis the space mean speed was used. In order to calculate the space mean speed the travel time was measured for all the vehicles which are to be considered for the study. The travel time was measured using Semi Automated Vehicle Analysis (SAVA), software developed at KTH by Jeff Archer (2006). In a similar manner to the micro study the vehicles which are under the impact of the red traffic light were excluded. In every five minute interval the vehicles were counted in order to know the flow. Basically, the vehicles under the green light were considered.. 20.

(23) Impact of Parking Maneuvers on Space Mean Speed And Average Travel Time 3.2.2. Regression Analysis. 3.2.2.1 Steps followed in regression analysis. According to Chatterjee and S.Hadi(2006) in their book ‘Regression Analysis by example’ here are some steps which need to be included in a regression analysis. Most of those steps are included in the regression analysis of this thesis work and these include: . Statement of the problem Selection of the relevant variables Data collection Model specification Choice of fitting method Model fitting Model validation and criticism Using the chosen model for the solution of the posed problem. The steps mentioned above are explained as follows, 1. Statement of the problem Usually problem formulation is the first step in regression analysis. Under this step formulation of the question(s) to be addressed by the analysis is included. Statement of the problem is the first and the most decisive part in regression analysis. Its significance comes due to the fact that wrongly defined problem will result in wastage of effort. In this thesis the main problem which is intended to be addressed is the impact of parking maneuvers. Normally there are various ways in which performance will be explained. Typically on this study the speed reduction due to the parking maneuver and related events will be analyzed. This means, the loss on the potential capacity of the urban roads due to some activities happening on the road side will be studied. This is a quantitative analysis on which the travel time is measured and the speed will be calculated as the distance is known. Travel time of vehicles vary due to various factors but for this study the situations in which travel time is affected by the factors chosen as an independent variable are considered.. 2. Selection of the relevant variables Selection of the relevant variables is the next step after defining statement of the problem. In this study the main aim is to see the impact of parking maneuvers on the traffic performance, double parked vehicles and the load/unload vehicles. For this purpose different category of events was developed. This includes the maneuvers, double parking and load/unload maneuvers. The maneuver embraces front inbound, 21.

(24) Impact of Parking Maneuvers on Space Mean Speed And Average Travel Time. reverse inbound and outbound maneuvers. In addition to these flow, the lane width and number of lanes are considered as independent variables. Besides these the bicyclists and pedestrians were included as relevant parameters. But on the preliminary analysis the two variables happen to be insignificant. Therefore, they are excluded. The dependent variable is the space mean speed and the independent variables are flow, maneuver, double parking, load/unload, lane width and number of lanes. 3. Data Collection After selecting the potentially relevant variables the next step is data collection. Sometimes data collection needs to be done in a controlled setting in the environment under the study to be used in the analysis. This is done to hold the effect of those factors which are not of primary interest. Usually, the data collection is carried out under nonexperimental conditions where very little is controlled by the researcher. In this study there is a video recording of 4hours and 30minutes in the morning and a 4hour recording in the afternoon. From this raw data the number of maneuvers under each category was counted. Besides this with the help of SAVA the travel time of all the vehicles which are not disturbed by the traffic light was measured. The number of vehicles using the street in each five minute segment was counted; with this the flow will be determined. 4. Model Specification The model that is suitable to analyze cause and effect is regression analysis. In this thesis the attempt is to evaluate the effect that comes on the speed due to numbers of variables. Therefore, multiple regression equation will be developed to deal with the situation. According to Chatterjee and S. Hadi on their book titled Regression by example, multiple regression equation is an equation containing more than one predictor variable. 5. Model Parameter Estimation The next step after the Model specification is model parameter estimation. There are various estimation methods. The most commons are ordinary least square method, maximum likelihood, the ridge method and others. Lots of statistical packages do the analysis. For this study Microsoft Excel will be used to make the model fitting, to estimate the regression parameters or to fit the model to the collected data. The estimates of the regression parameters (βo, β1, β2,….) are called the predictors. They are used to analyze effects of policy that involves changing values of the predictor variables or to forecast values of the dependent variable for a given predictor set up. With the help of regression analysis it is possible to understand the interrelationships among variables in a specific environment or circumstance. Sometimes the effect that. 22.

(25) Impact of Parking Maneuvers on Space Mean Speed And Average Travel Time. comes due to some covered values might be more valuable than the predictor variables on the equation. Model validation and criticism is done in the result part of the regression analysis. 3.2.2.2 Variable Definition. The regression analysis in this thesis work includes the following variables: . Space mean speed (dependent variable) Flow (Independent variable) Maneuvers (Independent variable) Double parking (Independent variable) Load/unload maneuver (Independent variable) Number of lanes (Independent variable). 1. Space Mean Speed The effect of different variables on traffic performance of an urban street is studied. In order to see the traffic performance space mean speed is selected as a representative parameter. In this study the impact of some events on the space mean speed was analyzed. For this reason space mean speed was selected as a dependent variable. The travel time which takes to cross a known distance segment was measured. Normally for this study the vehicles were selected in a controlled way. Vehicles could be disturbed from their normal driving condition for different reasons. Most of the factors which could make this delay are considered as independent variables. As it is explained on the description of the case studies, there is a traffic light at the end of the block. This condition brought a major challenge to select the proper vehicles for the study. In this study the delay due to the traffic light is not a major concern. Therefore one of the major factors to select vehicles is green traffic signal light. But besides this a long queue sometimes gets developed and a single green cycle will not be enough to dissipate the queue. Considering the entire mentioned criterion, the travel time was measured for all the vehicles. Then, an aggregate of five minutes was made for the whole day. All the measurements in five minutes were averaged to a single value. In this analysis there are 481 different observations. Each of these observations represents events in a five minute interval. Some vehicles enter the section in one five minute interval and leave the block in another five minute interval. Those observations are excluded from the data set to avoid consequent error.. 23.

(26) Impact of Parking Maneuvers on Space Mean Speed And Average Travel Time. 2. Flow The flow in an urban road is a dynamic factor which changes in every period of time. In this study one of the factors which are expected to affect the speed of a vehicle is flow. As it is an urban street segment, flow varies with time of the day. As it can visually be observed from the video of the site most of the queuing periods happen when the flow is high. For this study all the vehicles using the street segment are counted for every five minute interval. And finally the 481 observations corresponding to the averaged space mean speed are considered. 3. Parking Maneuvers The other most important variable which is considered to affect the speed of the vehicles is parking maneuver. It is the activity of vehicles to leave the drive way to the parking lane or join the moving traffic from the parking lane. Parking maneuver embraces various types of events; the forward inbound maneuver, reverse inbound maneuver, and outbound maneuver. These maneuvers have different impacts on the travel time or speed of moving vehicles, according to the visual inspection the inbound maneuvers happen to increase the travel time a bit more than the out bound. The reverse inbound maneuvers happen to take more time than the forward inbound maneuvers. For this reason their impact will also be relatively higher. Normally the presence of the bicycle lane on the street gives a good platoon for out bounding vehicles. The vehicles leave their position and wait for an acceptable gap on the traffic stream being at an angular position. Similarly, for inbound maneuvers specially to make a reverse maneuver some vehicles take aside and double park on the bicycle lane wait for an acceptable gap. For this study all the mentioned events were counted for every five minutes corresponding to the measured space mean speed. In some of the five minute intervals without any of such events a null value was assigned. 4. Double Parking Sometimes vehicles would like to stop at a specific location which is already occupied. This happens when delivery vans or trucks have to park in front of a specific store or sometimes when private cars have to drop passengers. Usually double parking takes place in the conditions that the driver is not going to leave the vehicle or if he is going to be in around. Taxis, delivery vans and delivery Lorries are the most common types of vehicle to make such a parking. Usually this type of parking blocks the bicycle lane so that the bicycles will be forced to mix with the flowing vehicles. In most of the cases the vehicles give way to the bicycles by reducing their speed. In such situation the vehicles will have a higher travel time. Besides this double parked truck or bigger size vans take some space from the traffic lane in addition to the bicycle lane. This will force the drivers to have a lower speed so that the travel time will be increased. For this study the presence of double parked vehicles will be checked and assigned 1 in every five minute 24.

(27) Impact of Parking Maneuvers on Space Mean Speed And Average Travel Time. observation. Some of the vehicles in the study stay in the double parking condition for more than five minutes; hence they happen to be in two time periods or more. And for the time periods without a double parking, a null value was assigned. In other words double parking is analyzed as a dummy variable. 5. Load and Unload Maneuvers These are maneuver of the vehicles which deliver goods for some stores and groceries. Normally, there is a special parking space for these vehicles. It is called Last Zone and is a region restricted only to the delivery trucks and vans. Usually it is a common practice to put a sign post for the time restriction and destination of the Last Zone. In addition to this the curb will be painted yellow. Parking other vehicles in this zone is forbidden and a passenger vehicle doing so is liable to a fine. Despite this fact some passenger vehicles park at these spots. This forces the delivery vehicles to make a double parking. This will affect the traffic significantly. In these special cases the load/unload maneuver will be taken as double parking instead. This event is not significantly different from the normal parking maneuver. The specialty with load/unload maneuver is that; it takes place at a specified location and it is an exercise of delivery vans and light trucks only. This independent variable is taken as a dummy variable. 6. Number of lanes The data set consists of one lane and two lane streets. Therefore, to capture the impact that comes due to the increase in the number of lanes, this variable is introduced. 7. Shock wave effect This variable is introduced to capture the effect of events in successive five minute time interval. It is introduced as a dummy variable. This means that if any type of events happen in the previous five minute interval it will be assigned 1 and 0 otherwise. All of the stretches taken for this study are not longer than 120 meter. So the impact of the events will be propagated to the adjacent blocks and subsequently to the next five minute interval. This situation will be captured by the shock wave effect variable.. 3.2.2.3 Expected Signs and Initial Hypothesis. From the entire description above, there follow some initial hypotheses which could be tested with the analysis of the model. These hypotheses are: 1. With the increase of flow the space mean speed is expected to reduce. For this reason the coefficient of flow is expected to be negative. 2. In the same manner when parking maneuvers happen the following vehicle always gets slowed down. From this it will be obvious that the parking maneuver is expected to have a negative coefficient. 25.

(28) Impact of Parking Maneuvers on Space Mean Speed And Average Travel Time. 3. As it is explained above usually the double parking events narrow the lane width and block the bicycle lane. For this and other reasons drivers will be forced to chose a much lower speed to drive safely in such streets. So double parking is also expected to have a negative coefficient. 4. Normally the load/unload maneuvers are the same as the other maneuvers except that they take place at a specified stretch of the street called Last Zone. Therefore, this variable is also expected to have a negative coefficient. 5. In multi-lane streets when various events happen the following vehicle will have enough space to escape from the disruption. For this and other reasons drivers will have a bit higher speed in multi-lane streets. As the number of lanes increase drivers choose higher speeds. Therefore, the number of lanes variable is expected to have a positive coefficient. 6. As it was explained in the variable definition, the light truck and vans which double park at any location along the street for the purpose of delivering goods to stores and supermarkets fall under the category of double parking. These vehicles are normally expected to have a higher reductive effect. The double parking variable is expected to have a higher magnitude than the other variables. 7. Because the dependent variable is space mean speed the intercept is expected to have a positive coefficient.. 3.2.2.4 Semi Automated Vehicle Analysis (SAVA) Software. SAVA software is a video analysis tool that is used for this study. On SAVA it is possible to measure the travel time and distance. The output from SAVA was used both in the micro analysis and macro analysis. Procedures in SAVA Micro Analysis -. -. Always the first step when dealing with SAVA is configuring the coordinates. To do these a known rectangular area is taken in the video. The dimensions and the diagonal of the rectangle are calibrated. Doing so the software will understand all the distances in the video. After the calibration it will be possible to measure distance at any location in the video. After calibrating the coordinates virtual lines are drawn with 10 meter interval. The time in the software is synchronized with the real time in the video recording. The time stamp at every 10 meter interval will be recorded to a text file output by the software.. Macro analysis -. The coordinates are configured in a similar manner as in the micro study. 26.

(29) Impact of Parking Maneuvers on Space Mean Speed And Average Travel Time. -. -. The time is also synchronized in the same way. Here the virtual lines are drawn at the entry and exit points. It is drawn at two locations (at the start and end of the specified segment which is the interest of study). The time stamp at the location was recorded for the vehicles of interest.. 4 Description of Case Study (Data Analysis) Basically three case studies are carried out. As it was already mentioned in the methodology part in Chapter 3, two approaches are used in this study: the Micro analysis and the Macro analysis. In the former one, three case studies are carried out. For the Macro study two different spots are used from each case study except the last one. For all of the spots an eight and a half hour study period is selected.. 4.1 Case Study One In this case study, two street segments are studied; they are named as Fleminggatan site one and Fleminggatan site two. Fleminggatan site one is the street segment which is found between Inedalsgatan and Polhemsgatan. This site has different activities and it incorporates special zones which are dedicated to the parking activity of specific vehicles. These include the last zone which is dedicated to only delivery vehicles and a restricted parking area for handicap cars. There is a hospital in this area and there are also many stores which increase the road side activity. The general road geometry is that it is a two lanes street which is situated between signalized intersections. The lane width is 3.3 meters and it has a bicycle lane which is 1.5 meter wide and located between the parking lane and traffic. The parking lane is 2.6 meters wide. On this street a 120 meter stretch was considered for the study. This site is one of the streets which lead towards the city center and it is a relatively busy street with a higher flow. On the road side of this street segment around 20 vehicles could be parked excluding the Last Zone.. 27.

(30) Impact of Parking Maneuvers on Space Mean Speed And Average Travel Time. Figure 4.1 Karta of Fleminggatan site one. Source: Google map. 28.

(31) Impact of Parking Maneuvers on Space Mean Speed And Average Travel Time. Figure 4.2: Road geometry of Fleminggatan site one 29.

(32) Impact of Parking Maneuvers on Space Mean Speed And Average Travel Time. Fleminggatan site two is the segment which is found between Wargentinsgatan and Norra Agenegatan. This site also has different activities. There is a last zone which is a reserved area only for delivery trucks and vans. The general road geometry is not somehow different from site one except that there exist some differences; it is a two lane street which is situated between signalized intersections. The lane width is 3.8 meters and it has a bicycle lane which is 1.5 meter wide and located between the park and the traffic. The parking lane is 2.6 meters wide. On this street a 140 meter stretch was considered for the study. This site is one of the streets which are coming from the city center and it is a relatively busy street with a higher flow. On the road side of this street segment around 22 vehicles could be parked excluding the Last Zone.. Figure 4.3: Karta of Fleminggatan site two. Source: Google map. 30.

(33) Impact of Parking Maneuvers on Space Mean Speed And Average Travel Time. Figure 4.4: Road geometry of Fleminggatan site two 31.

(34) Impact of Parking Maneuvers on Space Mean Speed And Average Travel Time. 4.2 Case study Two In this case study, one street segment with a variable lane width is studied. A street segment in Hornsgatan was studied. This segment is situated between Torkel Knutssonsgatan and Timmermansgatan. The specialty of this segment is that it is one lane for the first half of the stretch and the width increases significantly for the next half so that is will be considered as a two lane street. In this case study the first and second half of the street are dealt separately for the Macro analysis. The first half which is the one lane part incorporates a last zone in which load/unload activity will be carried out. This area is an active business area so that lots of parking maneuvers happen. Due to the high demand of the parking space numerous double parking events are visualized on the video. The general road geometry is that the first half of the street is one lane street. And it is situated past a traffic signal. The lane width is 3.0 meters and it has a bicycle lane which is 1.5 meter wide. The parking lane is 2.6 meters wide. A 55 meter segment was taken for the one lane part of the study. The other half of the segment has an average width of 5.2 meters so that it can be considered as a two lane. Normally a street width more than 5 meters is considered as a two lane street. It has a bicycle lane and parking lane which are continuations of the first half. This street has a moderate flow than the previous case study.. Figure 4.5: Karta of Hornsgatan. Source: Google map. 32.

(35) Impact of Parking Maneuvers on Space Mean Speed And Average Travel Time. Figure 4.6: Road geometry of Hornsgatan 33.

(36) Impact of Parking Maneuvers on Space Mean Speed And Average Travel Time. 4.3 Case study Three In this case study, a street segment which is found between Borgmästargatan and Ersta Trappor is studied. This site has different activities and it incorporates special zones which are dedicated to the parking activity of specific vehicles. These include the last zones which are dedicated to only delivery vehicles where the load/unload activities take place. The general road geometry is that it is a one lane street which is situated between signalized intersection and a pedestrian priority intersection. The total carriage way width is 5.9 meters. Of this width 2.6 meter is dedicated to the parking vehicles. Therefore, the lane width will be 3.3 meters and it has a bicycle lane which is 1.5 meter wide. Unlike all the other case studies the bicycle lane is located between the parking lane and the pedestrian side walk. For all the other case studies the bicycle lane is located between the traffic stream and the parking lane. On this street a 110 meter stretch was considered for the study. This site is one of the streets which lead towards the city center and it is a relatively busy street with a higher flow.. Figure 4.7: Karta of Folkungagatan. Source: Google map 34.

(37) Impact of Parking Maneuvers on Space Mean Speed And Average Travel Time. Figure 4.8: Road geometry of Folkungagatan. 35.

(38) Impact of Parking Maneuvers on Space Mean Speed And Average Travel Time. 5 Results and Discussion 5.1 Micro Analysis In the Micro analysis the data reduction is not done in a time bound manner. Some of the interesting events of the same kind are categorized together to see their impact. The interference of some other disturbing factors than the events of interest was more challenging to exclude. 5.1.1 Case Study One Micro analysis for case study one was carried out in fleminggatan site one. This site was the most challenging to get the data for this study. This street is highly busy and lots of maneuvers happen though. The problem is the street is occupied by queuing vehicles for most of the time. It was difficult to find vehicles which are affected only by a specified event of interest. Due to the fact that the street is so narrow and also there exists a long queue that generates from the red traffic signal simultaneous events happen most of the time. For this reason lots of vehicles were excluded from the data set. The results listed for this case study are somewhat indicative but should not be conclusive. 5.1.1.1 Free Flow Speed Trajectory. Average Travel time Trajectory Average travel time(seconds). 12 10 8 6 Travel time. 4 2 0 0. 20. 40. 60. 80. 100. 120. 140. Space(meters). Figure 5.1: Average travel time trajectory for every 10 meter segment. 36.

(39) Impact of Parking Maneuvers on Space Mean Speed And Average Travel Time. Space mean speed(Km/hr). Space MeanSpeed 50 45 40 35 30 25 20 15 10 5 0. speed. 0. 20. 40. 60. 80. 100. 120. 140. Space(meter). Figure 5.2: The space mean speed at every 10 meter space The average travel time and space mean speed are tabulated below space 10 20 30 40 50 60 70 80 90 100 110 120. Average travel time 0.98352941 2.04117647 3.09411765 3.94470588 4.73882353 5.65176471 6.52705882 7.42941176 8.29882353 9.17647059 10.1082353 11.1023529. speed 35.32548 36.28527 36.69526 42.05376 43.8492 41.85049 41.01211 40.8702 41.3971 40.99098 38.73615 38.07602. Table 5.1: Average travel time and space mean speed The results above give the average of 33 travel time and space speed of vehicles which are on a free flow condition. The graph for travel time appears to be perfectly linear. This tells that the time that takes the vehicles to travel every 10 meter segment is nearly equal. From Figure 5.1 above it can be seen that the space mean speed is around 40 km/h; it gets progressively increasing to some distance and falls again. This is because of the fact that the 37.

(40) Impact of Parking Maneuvers on Space Mean Speed And Average Travel Time. drivers come from a lower speed at the entry to the block and gets decelerated at the exit of the street. In the graph it can also be seen that the vehicles attain the maximum driving speed around half way the block and start decelerating. The average of the speeds at the 12 location is 39.76km/hr. So this can be taken as the free flow speed for this street. 5.1.1.2 Impact of Inbound Maneuvers. Impact of Inbound Manuevers on Travel Time 16 Avereage travel time(seconds). 14 12 10 8 Free flowing cars 6. Private cars. 4 2 0 0. 20. 40. 60. 80. 100. 120. 140. Space(meter). Figure 5.3: Average travel time trajectory for every 10 meter segment The inbound maneuvers happen to shift the graph upper than the free flow condition. The travel times are measured for 10 meter segments. The difference between the cumulated travel times, which is the average travel time at the 120 meter space, will give us the delay which comes due to the event; inbound maneuver in this case. As it was already explained in the methodology in Chapter 3, in the micro analysis the vehicles which are affected only with the inbound maneuver are considered. So these vehicles are typically affected by the event of interest. From the above graph, the private cars have an average delay of around 3.2 seconds.. 38.

(41) Impact of Parking Maneuvers on Space Mean Speed And Average Travel Time. Space mean Speed(Km/hr). Impact of Inbound Manuevers on Speed 50 45 40 35 30 25 20 15 10 5 0. Free flowing cars Private cars. 0. 20. 40. 60. 80. 100. 120. 140. Space(meter). Figure 5.4: The space mean speed at every 10 meter space. In Figure 5.4 above the private cars have an average speed lower than the free flow speed. As presented earlier, the average free flow speed is 39.76 km/hr. And the average of the space speed at the 12 locations happens to be 32.38km/hr for private cars when they are affected by the inbound parking maneuver. The space speed plot for the inbounding event affected vehicles happens to be in a harmonic way with the free flowing vehicles. As a result the gap between the graph lines is mostly constant.. 39.

(42) Impact of Parking Maneuvers on Space Mean Speed And Average Travel Time. 5.1.1.3 Impact of Outbound Maneuvers. Impact of Outbound Manuevers on Travel Time Average travel time(seconds). 16 14 12 10 8 Free flowing cars. 6. Private cars 4 2 0 0. 20. 40. 60. 80. 100. 120. 140. Space(meter). Figure 5.5: Average travel time trajectory for every 10 meter segment As it can be seen from Figure 5.5 above, the outbound maneuvers happen to affect the traffic in a same way as the inbound maneuver. As it was mentioned in the methodology part in Chapter 3, out bounding vehicles take enough time to have a good gap acceptance. This results in a lesser impact on the normal traffic. The average delay is around 4 second for the private cars.. 40.

(43) Impact of Parking Maneuvers on Space Mean Speed And Average Travel Time. Impact of Outbound Manuevers on Speed 50 Space mean speed(Km/hr). 45 40 35 30 25 20. Free flowing cars. 15. Private cars. 10 5 0 0. 20. 40. 60. 80. 100. 120. 140. Space(meter). Figure 5.6: The space mean speed at every 10 meter space As it can be seen from Figure 5.6 above the private cars have an average speed lesser than the free flow speed. The average space speed is 32km/hr for private cars. This implies a reasonable speed drop due to outbound maneuvers. As it can be seen from the plot the lowest point is around 30 meters from the start of the block this implies there exist a relatively higher amount of out bounding maneuvers at the specified location.. 41.

(44) Impact of Parking Maneuvers on Space Mean Speed And Average Travel Time 5.1.1.4 Impact of Double Parked Vehicles. Impact of Double Parking on Travel Time Average travel time(seconds). 14 12 10 8 Free flowing cars. 6. Private cars. 4 2 0 0. 20. 40. 60. 80. 100. 120. 140. space(meter). Figure 5.7: Average travel time trajectory for every 10 meter segment As it is shown in Figure 5.7 above, the travel time happens to be increased due to the double parked vehicles. Thus, the vehicles will be delayed by 2seconds due to the double parked vehicles. Normally, most of the double parking narrows the lane width. This will result in a higher travel time of the vehicles because of the fact that they choose a safe speed which is definitely lower than the normal free flowing speed.. 42.

(45) Impact of Parking Maneuvers on Space Mean Speed And Average Travel Time. Impact of Double Parking on Speed 50 Space mean speed(Km/hr). 45 40 35 30 25 20. Free flowing cars. 15. Private cars. 10 5 0 0. 20. 40. 60. 80. 100. 120. 140. Space(meter). Figure 5.8: The space mean speed at every 10 meter space The space mean speed of the vehicles happens to be lower than the space mean speed of the free flowing vehicles. The space mean speed for private cars is 35.74km/hr. The difference happens to be a bit lower when it is compared with that of the inbound and outbound maneuvers. When there is a double parked on the street the drivers choose lower speed than the free flow speed because the lane will be narrowed. Besides this the bicycle lane will also be blocked so that the bicycles also get mixed with traffic and the vehicles will be affected.. 43.

(46) Impact of Parking Maneuvers on Space Mean Speed And Average Travel Time 5.1.1.5 Impact of Load/Unload. Impact of Load/Unload on Travel Time Avereage Travel Time(Seconds). 16 14 12 10 8 Free flowing cars. 6. Private cars. 4 2 0 0. 20. 40. 60. 80. 100. 120. 140. Space(meter). Figure 5.9: Average travel time trajectory for every 10 meter segment As it can be seen from the graph above the load/unload maneuvers shift the travel time graph up. This is because of the fact that these maneuvering activities increase the travel time. The load/unload maneuvers increase the average travel time by 4seconds. Despite the fact that these maneuvers are similar to the other inbound and out bounding maneuvers, they have a bit higher impact. This comes due to some reasons: the first reason is the vehicles which are making the maneuver under this category are light trucks and Vans. And the other reason is the load/unload happens in the last zone only. So the drivers have to make sure that they are parking at the specified location. In doing so, they affect the following vehicles more than the other maneuvers.. 44.

(47) Impact of Parking Maneuvers on Space Mean Speed And Average Travel Time. Impact of Load/Unload on Speed 50 Space Mean Speed(Km/hr). 45 40 35 30 25 20. Free flowing cars. 15. Private cars. 10 5 0 0. 20. 40. 60. 80. 100. 120. 140. Space(meter). Figure 5.10: The space mean speed at every 10 meter space As is it seen in Figure 5.10 above, the space mean speed gets reduced with the load/unload vehicles. The space mean speed with the presence of load/unload vehicles is 31.7km/hr. This gives a significant speed reduction due to the load/unload vehicles. One significant thing that can also be seen from this figure is that, at 60 meter distance the speed drops significantly. On the road geometry, this location is where the last zone occurs. As it was already mentioned the specialty of this maneuver is the entire maneuvers take place at a specific location. Therefore the averages of all the impacts also manifest this fact.. 45.

(48) Impact of Parking Maneuvers on Space Mean Speed And Average Travel Time. 5.1.2 Case Study Two The data for the Micro and Macro analysis of case study two is collected from Hornsgatan. This street is also busy. As it was already mentioned on the description part this street is single lane with a lane width of 3 meters and gets wider to 5.2 meters after half way and could be taken as a two lane street per direction. This street is relatively good site for the study. The interference of red traffic signal is relatively less. On top of this the number of simultaneous events is less so that it is easy to take a specific event of interest without the interference of the other events. The draw back with this site is that there is no last zone in the direction of study. In this site it was easy and enough data was found to study the reverse inbounding maneuvers. 5.1.2.1. Free Flow Speed. Avereage travel time Average travel time(seconds). 12 10 8 6 Avereage travel time. 4 2 0 0. 20. 40. 60. 80. 100. 120. 140. Space(meter). Figure 5.11: Average travel time trajectory for every 10 meter segment. 46.

(49) Impact of Parking Maneuvers on Space Mean Speed And Average Travel Time. Space mean speed(Km/hr). Space mean Speed 50 45 40 35 30 25 20 15 10 5 0. Space mean Speed. 0. 20. 40. 60. 80. 100. 120. 140. Space(meter). Figure 5.12: The space mean speed trajectory at every 10 meter space Space. 0 10 20 30 40 50 60 70 80 90 100 110 120. Average travel time 0 0.956 1.852 2.716 3.518 4.32 5.142 5.964 6.764 7.648 8.574 9.56 10.544. Space mean speed. 39.11627113 41.38735078 42.48042168 45.99270783 44.60373067 45.39387671 44.42729233 44.7701295 41.35871816 39.97237216 37.63398698 37.41055392. Table 5.2: Average travel time and space mean speed The result above gives the average cumulated travel time and space mean speed of 25 vehicles. The graph of travel time appears to be perfectly linear. This implies that the time that every vehicle takes to cover all the 10 meter segments is nearly equal. From figure 5.2 it can be seen that for most of the time the mean speed stays around 40 km/hr. But it a similar pattern to case study one the vehicles start accelerating up to half way the block and start to decelerate. The average of the 12 locations is 42km/hr.. 47.

(50) Impact of Parking Maneuvers on Space Mean Speed And Average Travel Time. 5.1.2.2 Impact of Inbound Maneuvers. Impact of Inbound manueveres on travel time 16. Average travel time (seconds). 14 12 10 8 Free flow. 6. Inbound manuever affected 4 2 0 0. 50. 100. 150. space (meter). Figure 5.13: Average travel time trajectory for every 10 meter segment The inbound maneuvers happen to shift the graph to the upper left from the free flow condition. The travel times were measured in the same way as case study one for every 10 meter intervals. The vertical difference at the 120 meter location is taken as delay due to the event. As per this consideration the delay due to the inbound maneuver is calculated to be 3.2seconds.. 48.

(51) Impact of Parking Maneuvers on Space Mean Speed And Average Travel Time. Space mean speed( Km/hr). Impact of Inbound Manuevers on speed 50,00 45,00 40,00 35,00 30,00 25,00 20,00 15,00 10,00 5,00 0,00. free flow speed Inbound manuever affected. 0. 50. 100. 150. Space (meter). Figure 5.14: The space mean speed trajectory at every 10 meter space As it can be seen form Figure 5.14 above the graph for the vehicles affected by inbound maneuvers lies below the graph for the free flow condition. One interesting fact that can be seen from the graph is that in both cases the pattern is nearly similar. The gap between both the graphs is nearly equal. The average of the space speed at the 12 locations is 34.07 km/hr. As it was explained above the average free flow speed is 42km/hr.. 5.1.2.3 Impact of Reverse Inbound Maneuvers. Normally the reverse inbound maneuvers take more time than the front inbound maneuvers. As it can be seen in the video the vehicles take aside and double park in the bicycle lane till they find enough headway to make the maneuver. They have a bit higher impact than the front inbounding maneuvers. The delay due to this event is about 4seconds. In a similar way the space mean speed is also less than the case for front inbound maneuvers. The space mean speed for this case is 32.80km/hr.. 49.

(52) Impact of Parking Maneuvers on Space Mean Speed And Average Travel Time. Impact of Reverse Inbound Manuevers in travel time Average travel time (seconds). 16 14 12 10 8 Free flowing vehicles. 6. Reverse inbound. 4 2 0 0. 20. 40. 60. 80. 100. 120. 140. Space (meter). Figure 5.15: Impact of reverse inbound maneuvers on average travel time. Impact of Reverse inbound Manuevers in travel time Average travel time (Seconds). 50,00 45,00 40,00 35,00 30,00 25,00. Free flow Speed. 20,00 Reverse inbound manuever affected. 15,00 10,00 5,00 0,00 0. 20. 40. 60. 80. 100. 120. 140. Space (meter). Figure 5.16: impact of reverse inbound maneuvers in space speed. 50.

(53) Impact of Parking Maneuvers on Space Mean Speed And Average Travel Time 5.1.2.4 Impact of Outbound Maneuvers. Average travel time (Seconds). Impact of outbound manuevers in travel time 16 14 12 10 8 6 4 2 0. Free flowing vehicles Outbound manuever affected vehicles 0. 20. 40. 60. 80. 100. 120. 140. Space (meter). Figure 5.17: Impact of outbound parking maneuvers on travel time In a similar way to case study one the outbound maneuvers have lesser impact on the average travel time than the inbound parking maneuvers. As it can be seen from the graph the outbound maneuvers shift the graph from the free flow condition a bit. The delay is around 3.5seconds.. Space mean speed(Km/hr). Impact of Outbound Manuevers 50,00 45,00 40,00 35,00 30,00 25,00 20,00 15,00 10,00 5,00 0,00. Free fowing vehicles Outbound manuever affected vehicles. 0. 50. 100. 150. Space (meter). Figure 5.18: Impact of outbound parking maneuvers in space speed 51.

(54) Impact of Parking Maneuvers on Space Mean Speed And Average Travel Time. As it can be seen Figure 5.18 above the average space speed of the vehicles affected by the outbound parking maneuvers happens to lie below the free flow condition. And they are more or less following the same pattern. The average of the 12 observation is 32.87km/hr. This speed reduction is less than the one caused by the inbounding parking maneuvers. 5.1.2.5 Impact of Double Parking. Impact of double parking on travel time Average travel time (seconds). 16 14 12 10 8. Free flowing vehicles. 6 Double parking affected vehicles. 4 2 0 0. 20. 40. 60. 80. 100. 120. 140. Space (meter). Figure 5.19: Impact of double parking on average travel time As it can be seen from figure 5.19 above, the curve is shifted to the upper left. The double parking events increase the average travel time of the street. The average delay due to these events is around 4seconds.. 52.

(55) Impact of Parking Maneuvers on Space Mean Speed And Average Travel Time. Impact of Double parking on space mean speed 50,00 Space mean speed (Km/hr). 45,00 40,00 35,00 30,00 25,00. Free flowing Vehicles. 20,00 Double parking affected vehicles. 15,00 10,00 5,00 0,00 0. 20. 40. 60. 80. 100. 120. 140. space (meter). Figure 5.20: Impact of double parking on space mean speed As can be seen from figure 5.20, the space mean speed line for vehicles is affected by double parking lies below the free flow line. Throughout the stretch the shift is almost constant. And like the other mentioned events, the line follow almost similar pattern. Almost around half way the block the vehicles attain the peak speed. The average of the 12 points is 31.6km/hr. In a similar manner to the case study one the double parking happens to have significant effect than all the other events.. 53.

(56) Impact of Parking Maneuvers on Space Mean Speed And Average Travel Time 5.1.3 Case Study Three The data for the Micro and Macro analysis of case study three is collected from Folkungagatan. This street is also busy. This street is also a way for number of bus lines including the blue line. As it was already mentioned on the description part this street is single lane with a lane width of 3.3meters. This street is also good site for the study. The interference of red traffic signal is relatively less. The number of simultaneous events is less so that it is easy to take a specific event of interest without the interference of the other events. The specialty of this site is the bicycle lane is located between the outermost lane on which the vehicles make on-street parking and the pedestrians walk way. For this reason the drivers barley make double parking on this site. Therefore, enough data was not found for the analysis of the impact of double parking on this site. 5.1.3.1. Free Flow Speed. Avereage Travel time Avereage travel time(seconds). 12 10 8 6 Avereage travel time. 4 2 0 0. 20. 40. 60. 80. 100. 120. Space(meter). Figure 5.21: Average travel time trajectory for every 10 meter segment. 54.

(57) Impact of Parking Maneuvers on Space Mean Speed And Average Travel Time. Space mean Speed 50 Space mean Speed(Km/hr). 45 40 35 30 25 20. Space mean speed. 15 10 5 0 0. 20. 40. 60. 80. 100. 120. Space(meter). Figure 5.22: The space mean speed at every 10 meter space. space. 10 20 30 40 50 60 70 80 90 100 110. Average Travel time 0.956 1.852 2.716 3.518 4.32 5.142 5.964 6.764 7.648 8.574 9.56. Speed. 36.61414 39.22082 40.25884 40.22515 41.53121 41.44147 42.57149 42.93759 43.99609 43.95564 39.8731. Table 5.3: Average travel time and space mean speed The results above give the average of travel time and space mean speed of 27 vehicles which are on a free flow condition. The graph for travel time appears to be perfectly linear. This tells that the time that takes the vehicles to travel every 10 meter segment is nearly equal. From Figure 5.22 above it can be seen that the space mean speed is around 41 km/h; 55.

No results found