DECLARATION

I certify that except where due acknowledgement has been made, the work is that of the candidate alone; the work has not been submitted previously, in hole or in part, to qualify for any other academic award; the content of the thesis is the result of work which has been carried out since the official commencement date of the approved research program; any editorial work, paid or unpaid, carried out by a third party is acknowledged; ethics procedures and guidelines have been followed.

Requirements for writer I need to write, add, edit, proofreading and paraphrase.

Please see the following for the consideration:

Please use a reader-friendly font size (suggested is 12 points) and a line spacing of a little more than the standard 1.0 spacing

Please be careful with plagiarism, they use Turnitin.

ABSTRACT:  Writing between 250 -300 words.

ACKNOWLEDGMENTS:

• I would like to thank my senior supervisor, Dr. Gang-Jun Liu,….etc (writing 4 line)
• I am grateful and would like to thank the School of Mathematical and Geospatial Science…..etc(writing 4 line)

 

TABLE OF CONTENTS

TITLE PAGE …………………………………………………………………………………………………….. i

ABSTRACT ……………………………………………………………………………………………………… ii

DEDICTATION ………………………………………………………………………………………………. iii

ACKNOWLEDGEMENTS ……………………………………………………………………………….. iv

LIST OF TABLES …………………………………………………………………………………………….. v

LIST OF FIGURES ………………………………………………………………………………………….. vi

CHAPTER

CHAPTER 1: INTRODUCTION

• Introduction (writing, edititng and proofreading for this section) writing 2 pages

Cycling has increased in significance as a substitute mode of transportation. For instance, it is currently being used to address congestion, promote physical activities, address environmental issues like pollution, and improve public health (Iseki & Tingstrom, 2012).  In most of developed countries, bicycle-sharing programs have been set up and they continue to gain popularity. Cycling in Melbourne as a means of transport started in the19th century, nonetheless, mass car-ownership resulted into a dramatic drop in cycling in the 20^th century (Australian Bicycle Council 2004).

• Mention this chart With regard to Melbourne. Melbourne has shown very strong growth between 2006 and 2011, showing an increase of 1.6%. https://chartingtransport.com/

Source: charting transport(2013)

Cycling in Melbourne is used for commuting, recreation, and sport. There is an extensive network designed inform of off-road bicycle paths and lanes in the city. According to City of Melbourne (2015), the cycling network is comprised of more than 135 km of on- and off-road routes alongside with numerous on-street bicycle hoops. The mild climate and relatively flat topography have influenced bicycle riding in Melbourne. The rail trails accommodates all fitness levels and ages. Riders can choose from the different available options from fat rides to steep climbs. City of Melbourne (2015) has explained that most of these trails are covered with dirt or gravel, and they are appropriate for horses, mountain bikes, and walking. Touring bikes can be used around the city centre, which is mainly flat.

The study area chosen for this research is……….etc

This research sought to assess the……etc

• Statement Problem (edit and proofreading for this section)

The existence of bicycle paths, lanes, or facilities that are designated for safe usage by bicyclists. To cyclists, the bicycle paths and routes intertwine with some passing through steep and flat grounds. Therefore, they are compelled to use a lot of energy in instance where the slope is steep, which can cause accidents and crashes. Steep slopes are dangerous for bike riders, especially when going downhill, and they act as barriers when going uphill.  Longer slopes are supposed to be 2 per cent of less in grade (Bicycle Network, n.d).  Notably, grades more than 5 per cent should be avoided and used when necessary for a shorter distance. For bicycle riders, the use of routes with steep slopes and sharp bends are particularly dangerous. Therefore, addressing the problem through the use of GIS to establish the appropriate routes and determine slopes.

 

  1.3 Research Objectives and Research Questions

The research aim of the dissertation is to assess the use of GIS technology to analyse the impact of slope on bicycles riding. The objectives of this research include:

1. To use GIS technology to assess the impact of terrain slope (and hence path gradient) on bicycle riders.
2. To determine the slope for bicycle path segments.
3. To determine the shortest path that minimises travel time while taking into account the varying slope gradients of path segments.

The research questions are:

1. How GIS can be used to assess the directional slope gradient of the path segment, bicycle travel speeds and travel time?
2. How can uphill and downhill slope gradients be derived for a bicycle path segment?
3. How can the impact of slope gradients of path segments on bicycle riding be assessed between origin and destination within a study area?

 

CHAPTER 2: LITERATURE REVIEW (writing, edititng and proofreading for this section)  )

2.1 Introduction

            Bicycling is one of the most sustainable modes of transport and plays an important role in the transport system. Cycling has a number of benefits that include improving the quality of life, reduction of traffic congestion in cities, positive health outcomes due to the exercises involved, providing a suitable form of transport for short distances and environmental conservation among many other benefits. A number of cities have adopted bicycle master plans as they try to capitalize on the cycling opportunity (Litman et al., 2005). This has meant that bicycle ways are being created and existing roads are being redesigned to cater for cyclists. In trying to identify and build bicycle pathways, a number of factors are being considered including topography. A number of researchers have studied cycling trends and cyclist behavior and have found out that the slope gradient, travel time, travel distance, travel speed and energy consumption are factors that affect cycling. Research studies have revealed that bicyclists are more sensitive to topography and the slope. Slopes, especially steep slopes cause travel impedance. A number of researchers have come up with methods for slope-based bike segmentation such as Huang & Ye (1995). A number of methods to assess path segmentation difficulty due to slope have been developed and include their assessment ……………..etc, Some methods base on……..etc.

With advent of GIS technologies, it has become increasingly possible to map and plan for cycling routes, locating areas, determining the longest and shortest routes for cycling and map out hazards along the routes that may affect cyclists.

 

2.2 Importance of Bike Riding

Cycling has a large number of benefits as compared to the motorized forms of transportation of passengers. The benefits of cycling can be categorized into the health benefits, social benefits, economic and environmental and sustainability benefits. The first benefit of cycling relates to the easing of traffic congestion. Cycling greatly reduces the over-reliance on private cars and public transport (Teshome, 2006). Traffic congestion is a major issue in most urban areas and results in reduced productivity among the workers. As compared to the other forms of transport, cycling causes minimal congestion. Secondly, cycling has a wide range of health benefits. Among many cyclist, their motivation stems from health related reasons. Since it’s a physical activity, cycling translates to disease prevention and contributes to better public health outcomes (Winters, 2011; Teshome, 2006).

Additionally, cycling results in fewer emissions that cause pollution and fewer accidents as compared to the motorized transport (Fraser, SD & Lock, 2010).  Cycling reduces air pollution and results in the prevention of diseases such as heart attacks and obesity (Fraser, SD & Lock, 2010; Teshome, 2006). Cycling has a number of environmental benefits given that it is energy efficient, does not cause any air pollution and does not result in greenhouse gas emissions. Cycling greatly contributes to the mitigation of climate change which is a problematic issue in the world today. Due to the fact that it does not emit greenhouse gases, it ensures the protection of the earth. Cycling also provides a number of social benefits that include improving the social well-being of the community, provision of social capital and encouragement in community participation in issues (Teshome, 2006). Cycling is relatively cheap hence saving people the high transportation costs. It also provides a comfortable and fast mode of transport for the short journeys (Teshome, 2006). Some of the benefits are listed in the Figure below.

Figure  : The Benefits of Cycling

Source: Litman et al., (2005) (please explain the figure)

 

2.3 Factors Affecting Bike Riding (writing, edititng and proofreading for this section) 

  first write about general impact effecting bike riding , and then the slope definition and the impact of slope and other factor related to slope

           

2.3.1 Impact of Slope Gradient

Slope is defined by a plane at a tangent to a position on the surface. The primary components of a slope are gradient and aspect. The gradient is the maximum rate of change realized when a plane is elevated; while aspect is the direction of the place with reference to the arbitrary zero (Hickey 2000).  Slope length and angle calculations are significance when undertaking environmental analyses. To undertake the calculations, GIS technology has been applied because of its versatility and power for processing spatial data (Qiu, Esaki, Xie, Mitani & Wang 2006). The GIS is applied to provide a strong function for database processing and both in geostatistical analysis. Challenges in applying deterministic models to a regional slope has encouraged the use of GIS to obtain, check, and process  large spatial data sets, and solve geospatial related problems.

 

A huge body of literature reveals that there exist a number of factors that are relevant when addressing the impact of slope to bicycle riders and the suitable routes riders choose. For instance, factors such as the design of the street network, diversity establishments and the various land uses have a direct influence on the physical distance that cyclists have to cover from their points of origin to their potential destinations. Fraser & Lock (2010) found out that a number of unfavorable conditions found in terrain such as a steep slope, weather, traffic conditions, street connectivity, street density and the density of the road have a negative impact on the cyclists. For example, due to these conditions, the cyclists will be required to use more effort when they travel from one destination to another, which leads to an increase in travel impedance (Fraser & Lock 2010; Wardman, Parkins, & Page 2008). Besides, the presence of well-planned and improved facilities such as bike lanes, off-road bike trails, paths and sidewalks ensures a more comfortable ride for the cyclists. It has been found out that many bicycle riders are usually sensitive to the topography and will most likely avoid slopes that are steep and opt for the gradual and gentle slopes (Wardman, Parkins, & Page 2008). Many of the riders strive to use minimal human energy and reduce travel impedance (Wardman, Parkins, & Page 2008).

In a study conducted by Wardman, Parkins & Page (2008) in the UK, the researchers established that a slope variable was more significant as compared to all the other environmental variables. Similarly, a Delphi analysis in Iowa found out that a “mountainous topography” was the most significant factor during consideration for shorter paths and suitable routes (Souleyrette, Anderson, Hans, Mescher, Roeth & Thompson 1996). A hilly or a rolling topography, on the other hand, was found to be a factor of less significance to be considered by cyclists. According to Menghini et al. (2009), topography, street gradient, was a factor that was greatly considered by cyclists when choosing their routes. A combination of GPS and GIS analysis was applied and compared with the actual routes of the cyclists. Similarly, Celvero & Duncan (2003) found out that the decision to use bike paths was mainly influenced by the slope.

A number of factors are usually considered by cyclists when choosing routes to be used. Such factors include the slope, characteristics of the roads, environment, traffic and the route itself as shown in Table 1 below:

Table 1: Factors that Influence the Choice of Routes for Cyclists.

Source: Segadihla & Sanches (2014).

As it can be seen from the table 1 above, factors such as slope, road characteristics and pavements are very important to the bicyclists. Shankwiler (2006) established that many bicyclists preferred two-lane streets to wider roads. When it comes to the gradient of the road or the slope, it was established that the presence of an uphill stretch affects the route choice as the steep hills increases the effort of pedaling. It is for this reason that many cyclists avoid routes and paths with steep slopes (Menghini Carrasco, Schüssler, & Axhau, 2010). In their research, Stinson & Bhat found out those non-experienced cyclists preferred flat roads. On the other hand, the more experienced cyclists preferred steep slopes as they provided physical exercise. In a study by Broach et al., (2012) in Portland, Oregon, the researchers found out that the slope of the road was the most considered factor in choosing a road by a cyclist. Further, the researchers acknowledged that some of the more experienced cyclists were willing to ride 37% longer distances on flat routes so as to avoid slopes that are greater than 2% (Broach et al., 2012). Nonetheless, Winters, Brauer, Setton, and Teschke (2010) revealed that there was a lack of consensus with respect to gradient that is considered not suitable for bicyclists. A cyclist always aims for minimal travel distance. Also, a cyclist prefers minimal trip time as it implies the usage of little energy and effort (Haustein, 2013). An average cyclist would prefer safety over distance any time. Grade has been found to be another important factor of consideration by cyclists when choosing routes. For example, a steeper grade implies that a road is rarely used by cyclists.

2.3.2 Travel Time This factor should be related to slope after the impact of slope (Editing and proofreading)

Haustein (2013) established that minimal trip time is important to cyclists as it implies less effort and energy being used. Ellison & Greaves (2011) found out that the travel time factor influenced the cyclists’ decision for route choice and travel. Travel time has been found to greatly influence the competitiveness of bicycles and this competitiveness can be calculated using both empirical and hypothetical approaches. Sener, Eluru & Bhat (2009) further argued that travel time is very important to cyclists given that it is relative to the amount of distance travelled and the amount of energy used. On a similar note, Winters (2011) found out that a majority of the cyclists considered 30 minutes to be the optimum time for a cycling trip. A survey conducted in Texas by Sener (2009) established that the younger population (18-34) found travel time more effective as they preferred shorter distances of travel in terms of duration as compared to the adult population. Power must be considered when estimating the time and costs.

A study by Levisnson & Wu (2003) based on the Rational Locator Hypothesis which states that people are more likely to uphold a steady journey-to-work travel times as soon as they adjust their home and work distance, found that people were willing to cycle in as long as the distance was short and the travel cost was small. For instance, when the distance between Twin Cities was reduced, the amount of time spent when travelling reduced immensely. In a controlled geographical location, it was found out that the travel time increased from eight four minutes to ninety minutes. The conclusion made from the findings was that the time of travel remains the same when an individual rides a bicycle from one point to another in a flat road.

 

2.3.3Travel cost (this section not correct please rewrite  travel cost of what time or distance or direction  (writing, edititng and proofreading for this section) 

Literature on transportation economics and behavior of travel has established that bicyclists often make decisions about when, where and how to travel based on the concepts of generalized costs of travel and travel impedance ((Hanson 2004; Iseki & Tingstrom, 2012; Iseki & Taylor 2009). Further, a research study by Iseki and Tingstrom (2012) found out that travel impedance together with the generalized costs of travel reveal a number of issues that bicyclists often face such as the physical costs and the energy used in travelling. It is a common assumption that bicyclists are required to use the shortest routes possible in order to reduce the time of travel and the travel distance. Nevertheless, when considering travel time and travel cost, it is always important to factor in altitude change so as to effectively measure the distance between two locations in a flat surface. Further, Iseki & Tingstrom (2012) noted that most of the time data on travel time is usually not available and hence it is highly important to identify the factor that causes travel impedance in order to effectively and accurately estimate the generalized cost of travel by a cyclist.

 

 

2.3.4 Travel Speed    (writing, edititng and proofreading for this section) 

I used this source to calculate  speed from the slope according to Mike (2009). http://www.gis-t.org/files/Uz8W2.pdf

Please explain the formula (have a look to my presentation)

The velocity model used in this study made the following assumptions:

(slope is in decimal; i.e. a 1% slope is .01 decimal)

Base Velocity = 10 mph or 15 ft per sec

Uphill

• Velocity = 15 – [(+slope) * (125) ] ft per sec
• Minimum of 3 ft per sec (walking speed)

Downhill

• Velocity = 15 – [(slope)

* (500) ] ft per sec

• Maximum of 37 ft per sec (25 mph)

 

Parkin, J. and J. Rotheram. Design speeds and acceleration characteristics of bicycle traffic for use in planning, design and appraisal. Transport Policy, vol. 17, no. 5, 2010, pp. 335-341.

 

Slope has a significant effect on speed; when cycling downwards in a steep slope, the speed increases but decreases when cycling upwards. Additionally, more energy is needed when going up a slope as compared to when going down (Ellison & Greaves, 2011). Additionally, steep and long gradients are difficult to cycle up hence reducing speed. For example, travelling at a flat distance, the speed is usually 16km/h but this gradually increases, This increased the speed by 0.86km/h for each one percent of downhill gradient and decreased the speed by 1.44km/h for each one percent uphill gradient (see Figure 3)  16km/h + 5(0.86). But declines up the slope 16km/h – 5(1.44) as shown in Figure  below.

Figure 1: Effects of gradient on speed http://sydney.edu.au/business/__data/assets/pdf_file/0003/94926/ITLS-WP-11-06.pdf

Source: Ellison and Greaves (2011).  

 

 

2.4 Methods for slope based bike path segmentation ((writing, edititng and proofreading for this section)  )

Dividing geographic space into meaningful groupings is the basis of all mapping.  This holds for abstract maps like customer propensity clusters for purchasing a product or animal habitat suitability, as well as more traditional mapping applications such as vegetation maps, ownership parcels, highways and pipeline routes.  Further divisions of the patterns in a base map often involve segmentation.

A more complex form of segmentation is often referred to as “dynamic segmentation.”  This approach uses changes in conditions on other map layers to subdivide a route.  The Conditions-based segmentation procedure first derives a “universal conditions” map by overlaying a set of map layers to establish individual coincident polygons containing the same combination of conditions throughout their interiors (left side of figure 1).  A route is intersected with the universal conditions map and “broken” into a series of line segments at the entry and exit of each universal condition polygon.

 

The result is a series of variable length line segments with uniform conditions throughout their length.  The line segment set is written to a table containing x, y and z coordinates followed by fields identifying the combination of conditions along each segment.  In addition, the table can report “crossing counts” that note the number of intersections of the derived line segments and other linear features, such as river and road crossings.

 

Another advanced form of segmentation involves breaking a route into segments considering elevation changes along a terrain surface as shown in the bottom right portion of figure 1. Terrain-based segmentation divides the route into segments representing similar terrain configuration as determined by major inflection points and slope conditions.

 

The first step is to establish the elevation profile of the route by “masking” the terrain surface with the route.  To eliminate subtle changes, the elevation values are smoothed to characterize the overall trend of the terrain.  This is done by passing a moving-average window along the elevation profile resulting in a smoothed profile.

 

The width of the smoothing window is critical because if it is too large the smoothing will eliminate potentially important inflection points; if it is too small there will be a multitude of insignificant segment breaks.  The dilemma is similar to choosing an appropriate angular change factor in Deflection-angle segmentation and is as much art as it is science.  Experience for liquid pipelines suggests an 11 to 15-cell diameter window on a 10 meter digital elevation model (DEM) works fairly well.

Figure 2 illustrates the final step in the procedure.  The tool evaluates the smoothed elevation values by subtracting the current value from the previous location’s value as the solution progresses left-to-right along the smoothed profile.  If the sign is the same, a continuous “running slope” is indicated.  A negative difference indicates an upward incline; a negative difference indicates a downward descent; zero difference indicates no change (flat portion).  A location where the sign of the difference changes (negative to positive; positive to negative) identifies an inflection point.  The coordinates and elevation value for the location that changed sign is written to a Segments Table.

 

For example, the calculations in the upper left portion of figure 2 show a continued upward incline until the elevation point of 124 is encountered.  At this location the sign of the difference changes indicating the start of a downward descent (from a negative to a positive difference).   From: http://www.innovativegis.com/basis/BeyondMappingSeries/BeyondMapping_IV/Topic3/BM_IV_T3.htm

 

           

In their study, Huang & Ye (1995) established that bicyclists were more sensitive to slope as compared to motorists. The further noted that the conventional methods of calculating slope Arc/info fail to take into account the direction of the streets as they only give the surface slope.  Consequently, in making calculations, each street’s slope must be taken into consideration in collection of the necessary data. Huang & Ye applied the formula:

Where S_ab is the slope between the original point a and the final point b; h_a represents the elevation at point a; h_b the elevation at point b and l_ab is the distance between the two points a and b.

2.5 5.   Methods for assessing path segment’s level of difficulty for bike riding due to the impact of slope gradient (writing, edititng and proofreading for this section) 

Sources: http://www.gis-t.org/files/Uz8W2.pdf

Huang, Y & Ye, G., 1995. Selecting bicycle commuting routes using GIS.

http://www.tandfonline.com/doi/pdf/10.1559/152304010791232163

http://tigerprints.clemson.edu/cgi/viewcontent.cgi?article=2655&context=all_theses

http://www.tandfonline.com/doi/pdf/10.1559/152304010791232163

 

A number of methods to assess path segmentation difficulty due to slope have been developed and include their assessment based on Slope Gradient and

            Gradient, in cycling, refers to the steepness of a road section. A flat road has a zero per cent gradient while a road with a higher gradient for example 10% is steeper than one with a 4% gradient. Gradient greatly affects cycling and may cause travel impedance (Smith Jr, 1975). Roads with a gradient of over 5% difficult for riders while those with a 2% gradient are suitable for many riders (Litman et al., 2016). Riders have to contend with a number of guidelines that have been put forth to enable them ride easily and safely.

The International Mountain Bicycling Association (IMBA) has developed a trail difficulty rating system to guide planners in developing routes and cyclists in choosing routes to use. IMBA factored in the trail grade as an important aspect in assessing difficulty and found that the average grade for all cyclists to be 3%. A 3% grade can be used by both beginners, skilled, experienced and highly experienced cyclists (IMBA, 2012). Gicycle (2014) takes note of the fact that topography greatly affects the choice of routes and paths, and the fact that slopes affect speed and safety. In so doing they developed a model of assessing slopes and categorized slopes into the following categories: level, little slope, gentle slope, steep slope and very steep slope. A level slope and little slope are suitable for all cyclists since it is easy to ride. A gentle slope is suitable for riders with intermediate skills since it is more difficult. A steep slope is associated with riders with advanced skills due to its high level of difficulty. A very steep slope is suitable for riders with exceptional riding skills. Using this scale, bicyclists are able to choose the most appropriate route to use.

The Austroads Guide to Road Design on bicycle paths states that gradients that are steeper than 5% should be avoided unless it is a must to use them. The guide also takes note of the fact that an uphill rise for cyclists starts to become difficult at grades above 3% and become impossible at 5% grades and above. It goes ahead to state that grades of 3% are the maximum desirable grades for a cyclists path (Bicycle Network, 2016).

2.6 GIS based approaches to the measurement, assessment  and mapping of the impact of slope gradient on bike riding ((writing, edititng and proofreading for this section)  )

The Method of mapping  the impact of slope gradient

The of assessment  and mapping shortest path

            The shortest path and requires the least amount of travel time. Martello (2009) found out that travel cost, travel time and travel distance are the traditional impedances often applied in network analysis. These factors have an effect on the travelling degree. In Martello’s study, several origin and destinations were analyzed and the easiest path analysis conducted. The analysis is applied to determine the shortest path that is less sensitive to steep grades and minimizes time in selecting a path or route. The cost of impedance is considered during the calculation of the easiest path analysis. Huang & Ye (1995) argued that since impedance is not desirable, it is necessary for the path with the least impedance from origin to destination to be a representative of the most desirable route by cyclists from origin to destination.

 

The current bicycle models hold the assumption that the average speed along and among routes, and travel time is proportional to distance (Tokmylenko, 2013). Nevertheless, no method out there realistically determines the cycling time based on the speed change because of topography. A number of researchers have investigated the influence of topography on bicycle cycling and majorities are in conclusion that topography causes travel impediments, especially the steep slopes.

 

            Gradient and aspect are the primary components of a slope. According to Hickey (2000), gradient is the maximum rate of change that is realized on the elevation of a plane while aspect is the place direction with reference to arbitrary zero. In undertaking environmental analyses, calculations of the length of the slope and the angle are very important. GIS technology has widely been applied to calculate angle and length due to its versatility and its power to process spatial data (Qiu, Esaki, Xie, Mitani & Wang 2006). GIS also provides a strong function for the processing of data and geo-statistical analysis. The challenges encountered in applying deterministic models have encouraged the use of GIS technologies to obtain process and check large spatial data sets, and resolve any problems related to geospatial data. GIS is used in slope analysis and can be applied using a number of variables to determine bicycling in cities and towns. Moudon et al., (2005) sought to determine the relationship between cycling and the environment and considered variables such as distance to the neighborhood centers, land use, safety and how frequent people use buses. Other variables that were considered included the length of the sidewalk, traffic volume and speed, population density, topography and size of the street block. GIS was able to provide information on the frequency of bicycle rides in a week or month. GIS has and is producing information used in planning of the city.

Through GIS technologies we are able to select bicycle paths and routes. It also provides a wonderful opportunity to computerize the process of lane selection and routes. Huang & Ye (1995) embarked on a study to develop a method that could be used to select the most appropriate paths for bicyclists. The study revealed that GIS is an important tool in developing a data base that could be used effectively in route and path selection (Huang & Ye, 1995). Cyclists are more sensitive to surface quality and the slope. Application of GIS technologies in bicycle route planning requires more data layers as compared to vehicle routes, hence making it a different form of planning. After data collection using GIS, it was possible incorporate other valuable information from institutions and government agencies. GIS is now widely used to perform spatial operations such as terrain modeling and analysis of networks.

 

2.7 Conclusions (writing, edititng and proofreading for this section) 

The existing literature shows that the current methods being used to plan for bicycle infrastructure do not take into account the factors affecting cycling. Many planners often take into account economic feasibility and safety only, but a number of factors are being left out. As research reveals, factors such as the gradient slope, travel time and cost, length of routes and speed greatly affect bike riding. For instance, a steeper slope increases travel impedance as more energy and effort will be needed to go through it. GIS technologies have proven essential in bike riding as they are able to map out the locations hence enabling riders to move from place to place with ease. Additionally, with GPS and GIS, it has been possible to establish the easiest and shortest route for cyclists from one location to another. Besides, GIS analytics have also ensured that hazards along paths and routes are mapped out and hence enabling bicycle riders to avoid accidents.

 

 

 

 

 

 

 

 

 

 

Chapter 3 Research Methodology (writing, editing and proofreading for this section) 

• Introduction (writing)

 

 

 

 

 

 

 

3.2 Study Area editing and proofreading

The project study area is Rowville which is a suburb in Melbourne, Victoria, Australia, 27 km which is located in the south-east portion of Melbourne, Victoria.  Its local government area is part of the City of Knox. (See Figure 1). Rowville has a total land area of 19.5 km² (7.5 sq mi) and  from the results of the 2011 census the population of Rowville was 34,145 (ABS, 2011).

 

This project will analyse the Bicycle network within this study area to assess the effect of slope gradient on bicycle riding. The existing bicycle roadway network route system was examined using the variables of slope and distance in Arcmap 10.3, GIS to analyse and estimate the travel time.

 

 

3.3Data collection and preparation  (writing, editing and proofreading for this section) 

 

 

To assess and map the impact of path gradient on bicycle travel, we need to obtain the road network, mainly those used by the cyclists, a DEM of the Rowville suburb. The road network can be used to identify different types of roads, mainly those are used by the cyclists……etc

 

Table 3-1 Required input data and data sources

Data type	Data format	Description	Data source

 

 

.

 

The data sets on road infrastructure were sourced from Vic Maps. The DEM information was obtained from …

Hole data from https://services.land.vic.gov.au/landchannel/content/productCatalogue

 

 

In this study, the bicycle mode of travel is analysed for measuring travel impedance.

 

ArcGIS makes it possible to effecitvely and efficiently access, analyse and manipulate spatial datasets using feature datasets, geodatabases and feature classes. (Table 3.1).

 

Feature classes contain relational tabular datasets of spatial features that have identical spatial geometry (such as points, lines, or polygons), and data fields. The four most often utilised feature classes are polygon feature class, line feature class, point feature class and annotation (ESRI 2012).

 

A feature dataset contains feature classes that have the same coordinate system and that have related datasets. Feature classes can be thematically or spatially integrated using a feature dataset. The main purpose of feature datasets is to put together related feature classes into a single dataset for building a geometric network, a terrain dataset, a network dataset, or a topology, (ESRI 2012).

 

The model of geodatabase storage is built on a relational database structure and utilises a database management system (DBMS) datastore. Tabular datasets are utilised to retain the spatial geometry, base, constraints, and schema for each spatial dataset. Each row in the tabular dataset correlates to a single spatial feature. The shape attribute has information about the geometry of each spatial feature. SQL can be used to inquire, select and manipulate data from the feature classes, feature datasets, and geodatabases (ESRI 2012).

 

 

To measure the shortest route between two locations in Rowville, it is necessary to have a network point location dataset to be used as reference for shortest path analysis as from and to locations. For the purpose of this study, several such pairs of locations have been prepared mainly exemplifying some routes with interesting shortest path analysis results

 

 

3.4 Geoprocessing and analytical procedures

After all the datasets required for this study have been collected, prepared and

organized into a geodatabase, a number of geoprocessing and analytical procedures

have been developed to derive the outputs specified in section 3.5 that are required to

answer research question in section 1.2.

3.4.1 Mapping the bicycle network gradient                     Please mention the research questions to be answer it.

Determine the slope from road segment:

The formula below as applied by Huang and Ye (1995) is to be used to calculate slope.

 

Where S_ab represents the slope between initial point a and end point b, h_a is applied to present elevation at point a. in addition, h_b was the elevation at point b, while l_ab represents the distance between point a and point b.

 

The quantity in the numerator being the amount of vertical elevation change and the quantity in the denominator being the amount of horizontal position change.

To get the percent gradient or the percent slope I applied the formula as above and multiplied by 100.

The reason for applying the slope formula is that it provides a value for slope for every road segment.

 

 

<< INSERT FLOWCHART HERE >>

 

For the study, the following steps have been carried out to determine slope:

1. Split road line segments at vertices

Using the function Split Line at Vertices – it is possible to establish junctions at vertices of the network. This will cut existing road segments, at points of intersection in the network. At the end of the process, there are 7875 road segments. Before this geoprocessing, there were 1584. The newly created road segments will retain the relevant attributes of the original line segment where they belong.

 

2. Create road network SEQ_ID

Since there is no longer a unique attribute ID field that can be used in the road network dataset as a result of the previous splitting process, it is appropriate to create a unique SEQ_ID field within the road network dataset. Such a field can be used later on as a join key.

A field calculation is used to create the SEQ_ID. In ESRI support website there is a technical article for achieving this.  The URL is: http://support.esri.com/fr/knowledgebase/techarticles/detail/38517

Using the process described by that article, the following process was performed:

The code provided below will create a number sequence for data that has not been previously sorted. First, create a new integer field.  Then perform a field calculation on the new field, and set the Parser option to Python. Then choose Show Codeblock. We then put the following into the Pre-Logic Script Code (see figure 4):

 

1. rec=0
2. def autoIncrement():
3. global rec
4. pStart = 1
5. pInterval = 1
6. if (rec == 0):
7. rec = pStart
8. else:
9. rec += pInterval

return rec

 

Figure 4. The python prelogic script code for creating a sequential number field.

 

 

 

Then the following code is used for the actual field assignment:   autoIncrement()

 

 

3. Feature vertices to points

In order to have a clear point of reference for the slope calculations, a nomination of points to be used for reference is made. These points mainly consist of junctions and vertices of the road network dataset. The ArcGIS tool Feature Vertices to points is used to create the points.

This process is run twice–once for the From node, and the second time on the To node. This generates two datasets of points matching the number of road segments. The points created at this point retain the attributes of the road network including the SEQ_ID.

Subsequently these points can be used to get Z information from the raster.

 

4. Adding surface information to points

Using the tool “Add Surface Information” it is possible to add a Z attribute to the point dataset derived from the DEM elevation information. This tool will be run for both the From Node dataset and the To Node datasets. This geoprocessing tool will add a Z column data attribute which will be the elevation information from the raster at each of the point dataset locations.

 

 

5. Adding Z information to the road segments

Using a tabular join it is possible to add Z_From and Z_To attributes to the road segment dataset from the Z attribute of the related point dataset. Using the unique sequence field ID_SEQ created previously as join key the road segments and the point datasets From Node and To Node datasets can be joined. This allows the road segments to be populated with an additional Z_From and Z_To information obtained from From Node and To Node datasets respectively.

 

6. Calculation of slope

When the Z_From and Z_To information is available for the road segments, it becomes possible to compute the slope for each road segment using a field calculation.

A straightforward application of the slope formula will result in a slope value that can be positive or negative, depending on whether the segment is uphill or downhill. An absolute value slope can also be determined at this point.

 

 

7. Identification of slope classes

 

There is currently no precise way that slopes of road segments are quantified and measured. There is often just a vague indication for the level of slope or gradient (for example, “This road is far too steep.” or “This road has a gentle slope.”).

In order to reflect the slope in this model and to make it easier to understand the slope values, the following slope classes have been developed based on examples from literature review. See figure 5:

Verbal description	Gradient interval
Flat	0 – 1,5 %
Little slope	> 1,5 – 3 %
Gentle slope	> 3 – 6 %
Steep slope	> 6 – 12 %
Very steep slope	> 12 %

Figure 5. Slope classes and classification

The model used for the processing that was performed for this study is provided in figure 6

 

 

 

 

Figure 6: ArcGIS model builder to calculate the slope based on the road segments

 

7. Estimation of velocity

After the slope is determined, a formula obtained from Mike Martello’s 2009 study paper can be utilised to determine velocity or speed.

Reference URL: http://www.gis-t.org/files/Uz8W2.pdf

The reason for determining velocity or speed is that this quantity can later be used to derive the time impedance attribute value. The velocity attribute will give us an indication of the speed of bicycle travel along various road segments depending on the slope of the road segment.

The effect of the slope on speed can be explained based on the formula in figure 7.  What the formula will derive is that when the slope is flat, i.e. zero, then the speed will be the base flat surface speed. When the slope is uphill, then the bicycle rider will have a more difficult time in travelling. The steeper the uphill slope, the more difficult it will be. Therefore the formula used in this study will reflect this as well. The formula used has a minimum speed of travel for uphill slope. For uphill slope values that are too high the effect of the slope on speed is capped at the speed of walking speed–i.e. it is not possible to go slower than walking speed.

Conversely, when the bicycle rider is travelling downhill, the bicycle rider will accelerate. The formula employed for this analysis has a maximum speed, so that even if the slope downhill becomes far too steep, the derived speed value does not go beyond the maximum. This is perhaps reflecting reality too as bicycle riders will normally apply brakes in steep downhill situations so as to maintain a safe riding speed. The speed value thus derived from the formula will thus have a range of values depending on the slope, and it clearly reflects the correlation that uphill slope values result in slower speeds and downhill slope values results in higher speeds.

 

Base Velocity = 10 mph or 15 ft per sec

Uphill

• Velocity = 15 –  [(+slope) * (125) ] ft per sec
• Minimum of 3 ft per sec (walking speed)

Downhill

• Velocity = 15 –  [(­slope) * (500) ] ft per sec
• Maximum of 37 ft per sec (25 mph)

Figure 7. Speed formula from Mike Martello (2009)

 

The formula used indicates a linear relationship between slope and velocity. Some examples of values derived are provided below.

 

From_z	To_z	Shape_Length	Slope_Gradient	speed	assigned_spped
67.749934	67.307242	8.852952	-0.0500050153	43893.9515811671	40599.36
159.265572	166.068405	24.453178	0.2781983184	-21698.4813565911	3291.84
68.411524	70.130658	34.730351	0.0494994709	9669.8525661086	9669.8525661086
70.130658	68.411524	34.730351	-0.0494994709	43616.5897355658	40599.36

 

The conversion factor from feet to meters is:

0.3048

feet to m

 

The conversion of the flat surface speed values results in:

flat
15	fps
4.572	mps
16459.2	mph

 

The uphill minimum and downhill maximum speed values can be converted as well and the resulting values are:

uphill_min		downhill_max
3	fps	37
0.9144	mps	11.2776
3291.84	mph	40599.36

 

A python expression was used in field calculator to compute the speed. In the pre-logic code the following python script is defined to derive the Speed_From. The pre-logic code is below (figure 8):

def speed( s ):

base = 16459.2

uphillRate = 1371.60

downhillRate = 5486.40

minUphill = 3291.84

maxDownhill = 40599.4

 

if s > 0: # uphill

# Uphill

# • Velocity = 15  –  [(+slope) * (125) ] ft per sec

# • Minimum of 3 ft per sec (walking speed)  = 3291.84 m/h

 

speed1 = base – uphillRate * s

if speed1 < minUphill:

return minUphill

return speed1

else:

# Downhill

# • Velocity = 15  –  [(­slope) * (500) ] ft per sec

# • Maximum of 37 ft per sec (25 mph) = 40599.4 m/h

 

speed1 = base – downhillRate * s

if speed1 > maxDownhill:

return maxDownhill

return speed1

Figure 8. Python pre-logic script for calculating the speed from the sliope value.

The python expression used to assign the Speed_From attribute value is:

speed(  !Slope_Gradient!   )

 

The python expressions uses to derive the Speed_To values which is the speed of travel in the opposite direction is going to use the same expressions for the python field calculation. The only difference is in the assignment expression. The Speed_To assignment expression is:

speed(  -!Slope_Gradient!   )

This means that the negative or opposite direction of the Slope_Gradient value is used for the calculation. This implies that a slope that is uphill in one direction will be treated as downhill in the opposite direction and vice-versa.

The calculated speed will have units of meters per hour.

 

8. Calculation of trave time

 

After the speed is obtained it is possible to calculate a measure of the Time impedance required to traverse each road segment. The formula is based on:

distance = speed * time.

Which translates into:

Time = distance / speed

The units for distance or shape_length attribute is meters and the units for the speed attribute is meters per hour, therefore the computed time value is going to have units of hour. In order to translate the time values to units of minutes, the resulting values can be multiplied by 60.

There are two time attributes that will be derived: TimeFromTo and TimeToFrom. The VB expression for the field calculation for TimeFrom To is:

[Shape_Length] / [speedFromTo] * 60

The VB expression for the field calculation for the TimeToFrom (the opposite direction) will be similar to the one previously used but a different speed attribute is used:

[Shape_Length] / [speedToFrom] * 60

 

 

1. Mapping the shortest path

determines the realistic cycling travel time based on change in speed due to topography.

 

One objective of the project was to estimate shortest route while being sensitive to slope as an indication of degree of difficulty of tranversing the route segment. The shortest path is calculated on the basis of segment time impedance values previously derived.

 

The process of determining the shortest past requires the use of Network Analyst tool for solving the shortest path, and this tool requires a network dataset to be created first. The steps for creating this is the following:

 

9. Creating network dataset

Once the time impedance is computed, it is afterwards possible to pursue a network analysis of the dataset in order to eventually derive such calculations as shortest path based on impedance. In order to create the network dataset, a feature dataset needs to be created first.

Afterwards, the road network dataset is imported in the feature dataset. Then, a network dataset can be created on the road network.

 

10. shortest route analysis

Once the network dataset is built it is possible to perform shortest route analysis.

The overall view of the entire methodology can be seen in figure 8:

 

 

Flow Chart of the Procedure

 

 

3.5 Conclusion

It has been realised from this study that it is possible to perform an analysis of time of travel on a road network applied to bicycle travel and sensitive to slope road segment information. It can be seen from the results that the shortest route derived is not necessarily the same in both directions of travel between two locations. Certain road segments become more favourable as a travel route in terms of the shortest route analysis. This is reflecting the practical scenario where bike riders avoid steeper road segments if an alternative route is available.

Chapter 4. THE RESULTS AND DISCUSSION        (writing, editing and proofreading for this section)  Please mention the research questions to be answer it.

This section discusses how route total travel time differs based on a slope gradient sensitive network. The effects of slope gradient on cycling speed is also evaluated. The effect of slope gradient on shortest route analysis is also assessed. For this study, I ran two shortest (quickest) route scenarios where one assumed constant average speed and another accounted for speed difference based on the explained above. and proposed priority roads for infrastructure location.

 

1. Bicycle Network Gradient

It has been shown in this study that it is possible to derive slope gradient for road segments and this has been done using the slope formula. The slope formula is a very basic mathematical formula that derives slope as the amount of rise over the amount of run. It also follows from this that the slope is a unit less quantity.

 

In this study, the slope has been derived as both a signed quantity and an absolute value. Both attributes have been quite useful. The signed quantity is useful for scenarios requiring orientation and directionality. Such as during the derivation of speed, the evaluation of speed is needing to be sensitive to the slope orientation. When the slope is positive it means that the direction is uphill–e.g. there is a positive rise over positive run. When the slope is negative, then the direction is uphill–e.g. there is a negative rise over positive run.

 

The absolute value slope is useful for getting a general notion of steepness. Hence, the classification of slope in this study is done using the absolute slope. The absolute value slope gives the amount of the rise over the amount of run without the positive or negative sign–giving a result that is always positive.

 

It is only possible to derive the slope when the Z information of the bicycle road network segments is available. The Z information is necessary as this is the main determinant of the computation of the amount of rise in a bicycle road network segment. The Z information has been obtained from the start point and the end point of the bicycle road network segment.

 

The run amount is determined from the shape length of the bicycle road network segment. There is a limitation to this study in that the Z information is simply taken from the endpoints and then assumes that the slope derived from it is constant over the entire bicycle road network segment. This does not capture the fact that slopes on actual roads can gradually change over the length of the road.  This can be addressed somewhat by taking more road segments. Having more segments can result in a more precise evaluation of the slope. Of course, having more road segments can result in greater time for geoprocessing and GIS calculations, and for the purpose of this study a balance had to be made in the choice of number of road segments for analysis.

 

 

 

 

 

 

 

 

(Writing the discussion)

 

 

 

 

 

 

 

 

(Writing the discussion)

 

 

• Estimate travel speed

 

For this study, speed has been determined based on the speed formula from Mike Martello’s (2009) study, and it can be seen from the formula that the uphill and downhill direction of slope can have quite different effects on speed, and hence time of travel. The upshot is that the downhill slope can have a larger effect on speed and time of travel, than the uphill slope. The multiple applied on the uphill is only 125 ft per sec. The multiple applied on the downhill slope is 500 ft per sec. There is also a smaller range on the uphill speeds:  3 ft per sec to 15 ft per sec is the range of values possible on the uphill. The range for downhill speeds is 15ft per sec to 37 ft per sec.

 

 

 

For the purpose of this study, it has been necessary to simply assume that the speed is constant over the entire length of the bicycle road network segment. In reality, the bike rider will normally have episodes of acceleration and decceleration.  This is a limitation, as such an analysis will introduce a greater level of complexity.

 

Two speed values for each road segment: a FromToSpeed value and a ToFromSpeed value. This acts to distinguish the speed achieved on the two different directions of any particular bicycle road network segment taking into account the slope–i.e. uphill or downhill.

 

 

Figure 1  Portion of attribute table from bicycle road segment analysis table showing two different speed and time attributes per row or road segment entry

 

For this analysis, the time is derived as distance over speed. This study does not consider wait times. The time derived for a road segment has orientation, and the time taken to travel from point A to point B is not necessarily the same as the time taken to travel from point B to point A.  The time taken is affected by the orientation of slope–i.e. uphill or downhill. The time to go uphill is longer than the time take to go downhill over the same bicycle road network segment.

 

 

 

• Shortest path analysis: write the results and discussion

Analysis of shortest path has been performed on several pairs of origin and destination point locations. One such analysis result is shown below

 

 

 

 

 

 

 

 

(Writing the discussion)

 

(Writing the discussion)

 

(Writing the discussion)

 

 

Chapter 5. Conclusions ( writing 1 page)

 

 

 

 

 

 

 

 

Further Improvements (write 1 page)

1. Accessibility :………
2. Rider conditions…………………..
3. Path conditions………………..
4. Traffic conditions………….
5. Connectivity of bike network………………
6. Facilities for bike riding, parking and maintenance…………..

 

 

 

 

 

 

 

 

 

 

 

Bibliography check the sources and add new sources

American Association of State Highway & Transportation Officials., 2012. Guide
for the development of bicycle facilities (4th Edition ed.). Washington, DC: American Association of State Highway & Transportation Officials. Retrieved from from Pascal

Fraser, SD & Lock, K., 2010. Cycling for transport and public health: a systematic review of the effect of the environment on cycling. European. Journal of Public Health, vol.21, pp. 738–743

Forester, J., & Forester, J., 1994. Bicycle transportation: A handbook for cycling
transportation engineers (2nd ed.). Cambridge, Mass.: MIT Press.

Gray, M & Bunker, J., 2005. Kelvin Grove urban village regional accessibility by active transport. Queensland University of Technology. Print.

Gicycle., 2014. Slope-sensitive bicycle routing. Accessed April 17, 2016.  Retrived from https://gicycle.wordpress.com/2014/06/12/slope-and-routing/

Huang, Y & Ye, G., 1995. Selecting bicycle commuting routes using GIS. Berkeley Planning Journal, vol. 10, no.1, pp. 75•90

Iseki, H & Tingstrom, M., 2012, A New Approach in the GIS bikeshed analysis considering of topography, street connectivity, and energy consumption. National Center for Smart Growth Research and Education Urban Studies and Planning Program School of Architecture, Planning, and Preservation Available at:   http://www.wctrs-society.com/wp/wp-content/uploads/abstracts/rio/selected/3403.pdf

Litman, T., Blair, R., Demopoulos, W., Eddy, N., Fritzel, A., Laidlaw, D., Forster, K., 2005. Pedestrian and bicycle planning: A guide to best practices. Victoria Transport Policy Institute.

Nourian, G., Hoeven, F., Rezvani, S & Sariylidiz., 2015. Easiest paths for walking and cycling: Combining syntactic and geographic analyses in studying walking and cycling mobility. Space Syntax Laboratory,The Bartlett School of Architecture, UCL. http://resolver.tudelft.nl/uuid:02ff3640-b054-4ff7-8dbd-a4e99fadad5e

Smith, D. T. (1976). Safety and locational criteria for bicycle facilities: final report : prepared for Federal Highway Administration, Offices of Research and Development. Washington, D.C., Federal Highway Administration.

Teshom, E., 2006. A GIS-based bicycle suitability study in Addis Ababa:  Addis Ababa University. Retrieved from http://etd.aau.edu.et/bitstream/123456789/900/2/Eyob%20Teshome.pdf

Thistle, I., 2016. Build it: A Feasibility Study of GIS-Based Analyses of Cycling Infrastructure, Tufts University ,Available at:  https://sites.tufts.edu/MaryDavis/files/2015/09/Ian-Thistle-MA-Thesis.pdf

Tolley, R., 2003. Sustainable transport: Planning for walking and cycling in urban
environments. CRC.

US Department of Transportation., 2012. Traffic safety facts 2010. Washington, DC: US Department of Transportation National Highway Traffic Administration.

Wardman, M, Parkin, J & Page, M., 2008. Estimation of the determinants of bicycle mode share for the journey to work using census data. Transportation, vol. 35, no.1, pp. 93-109.

Winters, M. L. 2011. Improving public health through active transportation: Understanding the influence of the built environment on decisions to travel by bicycle. University of British Columbia https://circle.ubc.ca/bitstream/handle/2429/33377/ubc_2011_spring_winters_meghan.pdf?sequence=5

Yamashita, Y, Dantas, A, Taco, P & Yamamoto, K., 1998. Geographical information system & remote sensing in bicycle planning.

Zheng, F., 2011.  Modelling urban travel times. thesis. Delft University of Technology. https://www.google.com.au/url?sa=t&rct=j&q=&esrc=s&source=web&cd=1&cad=rja&uact=8&ved=0ahUKEwjfxo3Hrf_LAhXi26YKHe2iAl8QFggeMAA&url=http%3A%2F%2Frepository.tudelft.nl%2Fassets%2Fuuid%3Adada88aa-cf02-4ee9-9518-a8b9bdf1a654%2FDissertation_Zheng.pdf&usg=AFQjCNEnv_UcHmhiZVPocB1NbzLJJbKdXQ&bvm=bv.119028448,d.dGY

Identification of factors that influence cyclists route choice Source: Segadihla & Sanches (2014). http://redpgv.coppe.ufrj.br/index.php/pt-BR/producao-da-rede/artigos-cientificos/2014-1/891-identification-of-factors-that-influence-cyclistsa-route-choice/file