Evaluating public transport performance using System Efficiency Measure. The case of minibus taxi transport in Mekelle city

TABLE OF CONTENTS

ACKNOWLEDGMENT

ABSTRACT

TABLE OF CONTENTS

LIST OF FIGURES AND TABLES

List of figures

List of tables

CHAPTER 1 INTRODUCTION
1.1 General
1.2 Description of the study area
1.2.1. Demography
1.2.2Land use
1.2.3. Road network
1.2.4. Modal share
1.3. Problem of Statement
1.4. Objective
1.4.1. Main objective
1.4.2. Specific objective
1.5. Limitation of the Research
1.6. Beneficiary
1.7. Thesis Structure

CHAPTER 2 LITERATURE REVIEW
2.1. Introduction
2.2. Efficiency Indicators
2.3 system efficiency indicators
2.3.1. Public transport infrastructure availability
2.3.2. Quality of the Service
2.3.3. Productivity
2.3.4. Proximity
2.3.5. Customer satisfaction
2.4. Summary

CHAPTER 3 METHODOLOGY AND MATERIAL
3.1 Introduction
3.2 Data source
3.2.1Secondary data sources
3.2.2 Primary data sources
3.3 Sampling
3.3.1. Sampling of Respondents
3.3.2. Taxi sampling
3.4. Data collection techniques
3.5. Data analysis and interpretation
3.6. Presentation technique
3.7 Summery

CHAPTER 4 RESULT AND DISCUSSION
4.1. Introduction
4.2. Service productivity
4.2.1. Passenger per vehicle
4.2.2 Vehicle kilometer
4.3. Quality
4.3.1 Waiting time
4.3.2 Operational speed
4.4 Spatial coverage of the service
4.4.1Proximity to taxi route
4.4.2. Walking distance
4.4.3. The distance between routes and taxi stations
4.5 Infrastructure Availability
4.5.1 Road
4.5.2. Vehicle availability
4.6. Customersatisfaction
4.6.1Comfort
4.6.2. Accessibility
4.6.3. Safety
4.7Performance summery of the existing and proposal in relation to system efficiency parameters
4.8 Summary

CHAPTER 5 CONCLUSION AND RECOMMENDATION
5.1. Conclusion
5.2 Recommendation

REFERENCE

APPENDIX
Appendix1: population projection of Mekelle city
Appendix 2 Trip distribution calculation using the given data
Appendix 3 Data collected on the field
Appendix 4 Interview and questioner
Appendix 5 field data collection sheet

ACKNOWLEDGMENT

Thank GOD! Then, I would like to express my gratitude as below for the kind help of individuals that made the completion of the study possible.

First I would like to thank and appreciate my advisors Dr.AshenafiAregawi for the guidance and support he provided to finalize my thesis. I have benefited much from the discussions and consultations, and from the comments, I have received at every progress meeting. Again, he was fully committed and responsible throughout the completion of the study and for the guidance, he provided on the conception and development of the study.

Second, I would like to express my appreciation to officials of Mekelle road and the transport office for providing secondary data of all my data interests. Thirdly, I would like to thank Ms. Tdeg (from Mekelle road and transport), Mr. KbromHaregot and Ms. RigbeGebretnsa for their intensive data collection. I also would like to thank my colleagues from Mekelle University, road and transport chair for their support on providing necessary materials. Thanks are also for all the respondents who kindly participated in the study.

I would like to thank my family for their patience, and my sister Sara for the support she gave me from the conception to the documentation of this thesis.

Finally yet importantly, I would like to thank my husband Grmay Kahsay for his love and care; and I am heart fully thankful for all the suggestions, comments, motivations he provided throughout this thesis.

ABSTRACT

An effective evaluation of transport system efficiency is essential in establishing sustainable development of the transport system. Therefore, theperformance of taxi transport in Mekelle city has to be evaluated and modified to meet the objective of the city transport plan, which is providing safe, efficient, and affordable transportation to enable users to reach their destination at reasonable cost and time. Efficiency measures compare real and optimum level of input and output. In public transport, efficiency measuresuse to monitor resource and service distribution, identify deficiencies and opportunities.System efficiency parameters used to evaluate performance of taxi transport were infrastructure availability, customer satisfaction, quality, productivity and proximity.

Even though Mekelle is a rapidly growing city of northern Ethiopia, due to the limited extent of taxi transport, the inefficiency of taxi routes and the deficiency of taxi service people are spending more time and money on travel. Therefore, this thesis has investigated the current performance of taxi transport with regard to the efficiency parameters, identified deficiency of existing route network, and suggested improvements to the route network, and service deficiency.

To do so direct field data collection and questioner for primary data collection were used. ArcGISused to analyze the area coverage of the existing and proposed taxi routes under acceptable walking distance; AutoCAD used to draw the old and proposed taxi routes and taxi stations, Sigma plot, and MS-Excel used to draw the graphs. The analysis, discussion, and interpretation were done based on theories focused on public transport efficiency.

The analysis from the questioner and direct field data shows the problem with the existing taxi transport system in order are less infrastructure availability such as taxi route and minibus taxis, the available number of minibus taxi are serving only 13.18% of total trip and mode share of walking is 54.52%, unmodified proximity or less area coverage of the taxi systemwithin acceptable walking distance which is only 35%, long walking distance between taxi stations which is beyond acceptable range andpoor maintenance of roads. Based on those findings this thesis has calculated number of minibus taxi or buss required for different percentage of service increment and recommended buss with capacity of 80 passengers to be added. Even though buss with higher capacity can reduce congestion at times of low trip attraction and production passengers will be forced to loss more time inside the vehicles as the vehicles wait until they reach their capacity. So government has to subsidize the sector in order to make the vehicles move even with less passengers to reduce the passenger time lost in side vehicles and make the system more efficient.It has also proposed new taxi stations and routes to improve the area coverage and to reduce the walking distance from one station to another. In addition, the thesis has also proposed the routes to be maintained and added in priority.

Keywords:proximity, Route network, System efficiency, Taxi transport

LIST OF FIGURES AND TABLES

List of figures

Figure 1: Mekelle sub-city boundaries

Figure 2: Map of existing land use

Figure 3: Existing road network of Mekelle city

Figure 4: passenger per vehicle of each route

Figure 5: vehicle Kilometer in each route

Figure 6: Taxi waiting time of individuals

Figure 7: Route number versus average speed in each route

Figure 8: Segment pictures on Enda Gebriel and Adi-Da'ero taxi route

Figure 9: Area coverage at 500m walking distance from current taxi route

Figure 10: Walking distance of individuals

Figure 11: Existing taxi routes in Mekelle city

Figure 12: Proposed taxi routes

Figure 13: Area coverage at 500m walking distance from proposed taxi route

Figure 14: Existing taxi station

Figure 15: Picture on taxi stations around Kedemay Weyane

Figure 16: Proposed taxi stations

Figure 17: Current taxi distribution in each sub city

Figure 18: Socio economic characteristic of samples in questioner

Figure 19: Customer expectation from a modified system

List of tables

Table 1: Population distribution of sub cities

Table 2: Trip attraction of different land use

Table 3: Land use components and area coverage of each land use

Table 4: Development of road building in Mekelle city in different year

Table 5: Modal share of different modes in Mekelle city

Table 6: Summery of data collected from municipality

Table 7: sample distribution of respondents for each sub-city

Table 8: vehicle sample distribution of each sub city

Table 9: Methodological framework

Table 10: Taxi distribution in each sub-city

Table 11: Trip generation of Mekelle city for different years

Table 12: Trip distribution percentage of each sub-city

Table 13: Trip production and attraction of each sub-city

Table 14: Travel time versus friction factor

Table 15: Fij for all sub-cities of Mekelle city

Table 16: O-D matrix

Table 17: Modal split of Mekelle city

Table 18: number of vehicles required to be added for different percentage of service increment in each sub city

Table 19: Summary of customer satisfaction

Table 20: summery of the existing and proposed system in relation to system efficiency parameters

CHAPTER 1 INTRODUCTION

1.1 General

This chapter starts by presenting the problem with current urban transportation in developing cities. It then briefly describes the study area. It defines the research problem, research objectives and in the end, it explains the thesis structure.

Transportation is a movement of people and goods from one point to another. Transport plays a great role in social and economic activities so providing efficient transport is mandatory for cites especially in the modern era where cities are the center of urbanization and global economies (Bizunesh, 2016).

In recent years despite the unbalanced demand and supply of the urban transport system, due to restrictions on land resources and environmental capacity, the number of motor vehicles in cities is increasing with economic development and urbanization. This is causing traffic congestion in urban areas (Jin Qin 1, Yuxin He 1 and Linglin Ni, 2014).

Public transport is a name given to different modes that are available to the public irrespective of their ownership. Modern public transport modes are classified into 4 categories based on their operation, those are Buses and trolleybuses, Light rail transit Trains, Rapid rail transit (Metro, subways or underground) and Sub-urban rail transit (Commuter rail system)(Bizunesh, 2016). Even though it can be provided by different modes of transport in most developing countries it is provided by busses and mini buss(NIGER, 2011). In Mekelle city, public transport is provided by minibus taxi.

Public transport is often faster and cheaper than using a car especially in congested areas where parking is limited. The use of public transport also has an impact on our environment and communities, it causes less pollution and congestion and enables a more active and connected community. Public transport helps to reduce transport impact on employees and reduce the need for parking lot on the worksite; it also helps employees reduce their travel cost. Public transport also has benefits such as reducing the cost of driving and maintenance reduces carbon emission, it avoids driving stress and parking requirement, it can improve health by leading to different exercises and helps to use travel time for doing other activities such as reading (Falconer, 2013).

In Sub-Saharan countries such as Addis Ababa and Accra, the focus given to urban transport modes of poor people in providing affordable forms of public transport is less. Even though almost 70% of the person trips depend on public transport in large urban areas, it is neglected by the government, poorly financed and poorly managed (A. Kumar,2012).

Mekelle city lacks smooth mobility, standard and futuristic city road network. Existing roads are narrow and crowded by the private companies, which tend to build high-rise buildings. The city contains less than 30 kilometers of asphalt roads connecting the city within itself. In addition, because of that, mobility-related problems have become common in the city (Mesele A and Tsehaye Z, 2016).The existing public transport service in Mekelle, minibus taxi and Bajaj, fails to fulfill the demand. Excessively long waiting times as well as long walking distance characterize the current taxi transport situation in Mekelle city and this is worst in morning and evening hours (Menasbo Gebru and Desta Kidanu, 2013). Even though public transport demand of Mekelle city is high the existing taxi transport supply and demand is unbalanced due to different causes such as poor infrastructure quality, affordability, etc (Menasbo Gebru and Desta Kidanu, 2013).According to transportation plan, implementation strategies, and land acquisition of Mekelle city,theobjective of the citywide transport plan is to provide safe, efficient, and affordable transportation to enable users to reach their destination at reasonable cost and time(Mekelle city transport office, 2011).

Public transport is not responsive to change in demand and supply to like the private vehicle. So the need to make public transport responsive to change is high(NIGER, 2011).So the need to make studies on the current public transport is high to create responsive public transport.

Efficiency measures are tools that can help transport officials to plan the future transport system and how to get there(Chhavi and Dhingra, 2011).Therefore evaluating the existing taxi transport performance using efficiency measures is essential to identify the major areas where to take action to modify the system.

1.2 Description of the study area

The study area, which is Mekelle city, has geographical area coverage of 191.51 square k. meters.Mekelle is divided into seven sub-cities namely Hadnet, Adihaki, KedamayWeyane, Ayder, Semien, Hawelti, and Quiha as shown in the map below. The objective of such administrative structural rearrangement was to address the public demands and to enhance equity, effectiveness, and efficiency of service(Mekelle city road and transport office, 2014).

illustration not visible in this excerpt

Figure 1: Mekelle sub-city boundaries source: Mekelle city administration

1.2.1. Demography

Demography with land use has a direct link with the trip production and attraction so transport plans have to consider the demography and land use. The table below shows the population distribution of each sub city and their density. As shown in the table Hawelti sub-city is the most dense sub city in the Mekelle. It inhabits 20.8% population of the city in 5.8% of area coverage.

Table 1:Population distribution of sub cities

illustration not visible in this excerpt

1.2.2 Land use

Land use highly affects transport demand so transport modeling has to incorporate land use.Land-use planning has a key role in providing an integrated transport strategy.Planningwhich integrates the location, scale, density, design, and mix of land uses can help reduce the need to travel, reduce the length of journeys and make it safer and easier for people to walk, cycle or use public transport. A decrease in urban pollution can be achieved through better land use development and better transportation plan. The relation of land use to the trip attraction of Mekelle is as shown in the table below. Areas such as center of the city, commercial areas and areas with administration offices have high trip attraction therefore higher transport vehicles are required to be assigned to such places.

Table 2: Trip attraction of different land use

illustration not visible in this excerpt

Source: Mekelle city road and transport office

Table 3: Land use components and area coverage of each land use

illustration not visible in this excerpt

This image was removed due to copyright reasons.

Figure 2: Map of existing land use

1.2.3. Road network

Streets in the old city which is around Kedamay Weyane are crowded in addition to this street parking is also contributing to the congestion. The road share of the total area of the city is 10% which is way less than the standard share (Mekelle city administration, 2014).

The streets stop functioning during heavy rain due to the lack of proper drainage. This contributes to the congestion increment at the center. The blocks on the old city center are short and streets lack hierarchy plus, some trees on the two-way street block the visibility and this makes driving dangerous. Several upgrading of roads is underway. Improvements has been achieved after the introduction of cobblestone but some cobblestone roads luck the required qualifications such as surface irregularity and gaps among the cobblestone pieces, which makes them unsuitable to drive and lucks accessibility to disabilities. The level difference between the built-up structures and the road lead the owners to build an informal staircase on the entrance of the buildings consuming the walkways of the pedestrian. In addition to this pedestrian walkway give access to shopping and selling goods squares are also functioning for recreation and sporting activities and shopping activities. Designs do not consider this situation (Mekelle city administration, 2014).

The city lacks smooth mobility; existing roads are narrow and crowded by the private companies, which tend to build high-rise buildings. The city contains less than 30 kilometers of asphalt roads connecting the city within itself, because of that, mobility problems are affecting the life of the people in the city(Mesele A and Tsehaye Z, 2016).

The existing public transport service in Mekelle, minibus taxi and Bajaj, fails to fulfill the demand. Excessively long waiting times as well as long walking distance characterize the current taxi transport situation in Mekelle city and this is worst in morning and evening hours(Menasbo Gebru and Desta Kidanu, 2013).

public transport demand of Mekelle city according to the previous study is high since the existing taxi transport supply miss much the demand due to different causes such as poor infrastructure quality, affordability, etc(Menasbo Gebru and Desta Kidanu, 2013).

According to transportation plan, implementation strategies and land acquisition of Mekelle city objective of the citywide transport plan is to provide a safe, efficient, and affordable transportation to enable users to reach their destination at reasonable cost and time(Mekelle city transport office, 2011).

Table 4: Development of road building in Mekelle city in different year

illustration not visible in this excerpt

Source: Mekelle City Administration

This image was removed due to copyright reasons.

Figure 3: Existing road network of Mekelle city

1.2.4. Modal share

The main mode of transport is walking, even though the pedestrian walkway facilities are very poor. The level difference between the built-up structures and the road lead the owners to build an informal staircase on the entrance of the buildings consuming the walkways of the pedestrian. In addition to this, pedestrian walkway gives access to shopping and selling goods. Squares are also functioning for recreation, sport activities and shopping activities(Mekelle city administration, 2014). According to the same reference the share of different modes in the city is as shown in the table below. The rest which is 58% of the trips are made by walking.

Table 5: Modal share of different modes in Mekelle city

illustration not visible in this excerpt

1.3. Problem of Statement

Mekelle is the rapidly growing city of Northern Ethiopia with dynamic expansion. However, people are spending more time and money on travel as the current taxi transport is characterized by excessively long waiting time and long walking distance. Therefore, network and service deficiency has to be investigated to identify the extent and recommend possible solutions on improving the service.

1.4. Objective

1.4.1. Main objective

The main objective is to investigate Mekelle city’s taxi-transport performance using system efficiency parameters, toidentify route network deficiencies, and make recommendations on possible improvements.

1.4.2. Specific objective

- To investigate the performance of existing taxi transport using system efficiency parameters such as infrastructure availability, customer satisfaction, quality, productivity and proximity.

- To identify deficiency of existing route network

- To suggest possible improvements to the route network

1.5. Limitation of the Research

This thesis studies as Mini Bus Taxi are the only solution for public transportation. It doesn't consider the other public transports such as Amora City Bus because they don’t give service regularly. And also it doesn’t take in to account the Bajajs, as they are controlled by their owners and they are engaged in private contract ride service rather than working in public transport with the estimated tariff and delineated routes according to the interview with transport officials.

1.6. Beneficiary

The taxi unions can have an awareness of the spatial coverage of the service.And helps them to identify where to take action to improve the service of the system.The future city master plan designers will have an initiation on the coverage of the service and lead them to develop the master plan in a way that can improve the stated problem. And it will give further research opportunities.

1.7. Thesis Structure

Chapter 1: This chapter starts by presenting the problem with current urban transportation in developing cities. It then briefly describes the study area. It defines the research problem, research objectives, and thesis structure.

Chapter 2: Defines efficiency, system efficiency and defines the parameters used for measuring the efficiency of public transport based on a literature survey.

Chapter 3: This chapter deals with the methodology in which the data source, data collection, and analysis techniques are discussed.

Chapter 4: Describes the current situation of taxi service based on the data collected for the efficiency parameters defined in chapter 2 mainly using statistical analysis and GIS and summarizes the overall performance.

Chapter 5: This chapter proposes a new route and new taxi station based on the result found from the previous chapter, evaluates the proposed route and station based on the efficiency indicators provided in chapter two, and compares with the current route and taxi station.

Chapter 6: Gives a conclusion based on the analysis result and recommends improvements for future.

CHAPTER 2 LITERATURE REVIEW

2.1. Introduction

White, (2002) Defines public transport as all modes of transport, which are available to the public. Public transport provides efficient means of moving a large number of people to achieve the demand in a city(NIGER, 2011). Public transport is often faster and cheaper than using a car especially in congested areas where parking is limited. The use of public transport also has an impact on our environment and communities, it causes less pollution and congestion and enables a more active and connected community. Public transport helps to reduce transport impact on employees and reduce the need for a parking lot on the worksite,italso helps employees reduce their travel cost(Falconer, 2013).

In Africa public transport is provided by low capacity vehicles; such as minibus taxis and it is operated by private sectors this leads to the requirement of large number of vehicles to satisfy the demand. Minibus taxi is vehicle with a capacity of 5 up to 35(Africa, 2010). In Mekelle, public transport is provided by minibus taxi owned by private.

“Public transportation is a relatively high-capacity and energy-efficient alternative for urban passenger transportation, as compared with the private automobile” (Yali Yang,Hao Chen and Lihua Chen, 2010).

The basic requirements of public transport are safe, efficient, attractive, affordable and sustainable (NIGER, 2011). Therefore, efficiency is a basic requirement of public transport.

Efficient public transport is a transport system with a clear network, proper organization, and management, competitive market, and sound finance mechanism (NIGER, 2011). Efficient public transport can serve as an environmental safeguard for conserving energy, protection community quality of life, and facilitating urban economic growth and development.

Efficiency/performance evaluation of the transport network is essential to establish sustainable development in any transportation system. A quantitative evaluation reflects the effect of network structure, traffic demand, travel choice, travel cost on network efficiency(Jin Qin 1, Yuxin He 1 and Linglin Ni, 2014).

There are close to half a million population living in Mekelle city. The city Lacks smooth mobility, standard and futuristic city road network does not exist, at least in a visible way to the public. Existing roads are narrow and crowded by the private companies, which tend to build high-rise buildings. The city contains less than 30 kilometers of asphalt roads connecting the city within itself. Because of that, mobility problems are affecting life in the city(Mesele A and Tsehaye Z, 2016).

Even though the objective of the citywide transport plan is to provide safe, efficient, and affordable transportation to enable users to reach their destination at reasonable cost and time(Mekelle city transport office, 2011).The existing public transport service in Mekellefails to fulfill the demand. Long waiting times as well as long walking distance characterize the current taxi transport situation in Mekelle city and this is worst in morning and evening hours(Menasbo Gebru and Desta Kidanu, 2013).

“Efficient public transport can attract economic activity to cities, and boost productivity by improving connectivity and reducing time lost to travel”(Siemens, 2017). Efficient transport has a great role in improving the quality of life by making the city more attractive to live in and helps people to deliver their products and consume what they need easily.

Jin Qin 1, Yuxin He 1 and Linglin Ni, (2014) stated that efficiency/performance evaluation of the transport network is essential to establish sustainable development in any transportation system. Therefore, evaluating the efficiency of public transport is mandatory to identify the opportunities and deficiencies with the system to take systematic action on making the service most efficient and effective.

2.2. Efficiency Indicators

There are different efficiency categories such as system efficiency, network operating efficiency, labor efficiency, utilization efficiency, and finance efficiency(Abreha, 2007). There are five main components in assessing public transport efficiency those are system efficiency, network efficiency, institutional framework efficiency, organization efficiency and finance efficiency(NIGER, 2011).According to Abreha,(2007) indicators of system efficiency are infrastructure availability, quality, proximity, productivity, and customer satisfaction. To find system efficient public transport factors such as special, public transport demand, network, and infrastructure have to be included.

2.3 system efficiency indicators

2.3.1. Public transport infrastructure availability

Infrastructure in public transport is the vehicle that is used to transport passengers irrespective of their possession and the road they use when they travel from one origin to destination(White, 2002).

- Vehicle availability: The vehicle infrastructure determines the maintenance and operating cost. It determines the speed and the loading capacity of the system.
- Road: The road infrastructure is the available route whether the traffic is mixed, where public transport computes for space, or bus only lane, where public transport has its own lane to use.

2.3.2. Quality of the Service

Acceptable Quality of service varies from a country to another. Service quality is affectedby different criteria such as income level, availability of alternative mode of transport, geographic and climatic condition, people’s attitude and ethnic characteristics, the value of time, traditional standards. Even though there are no set of standards to measure the quality of service in a particular country there are attributes in a service that can be measured such as waiting time, walking distance and average network speed(Alan Armstrong -Wright and Sebastien Thiriez, 1987).

Waiting time: is a time where passengers have to wait for a taxi at a taxi station or on the route. It is a primarily perceived service quality indicator by the users. The overseas unit of transport and road research laboratory (TRRL) sited in Alan Armstrong -Wright and Sebastien Thiriez, (1987) have concluded that in developing countries the average waiting time should be 5- to-10 minutes with a maximum waiting time of 10- to-20 minutes.

Walking distance: is the distance that passengers have to travel to a taxi station or from the station to their destination. It is also an indication of area coverage. I well-served urban passengers are expected to find public transport in 300-500 meters from their home or workplace. In low-density areas, walking distance above 500 meters is acceptable but it should not exceed 1000m(Alan Armstrong -Wright and Sebastien Thiriez, 1987).

Average network speed: weighted average by the number of vehicles provided in each route. The network speed is the ratio of the route length to travel time. Route length is the distance covered from a particular origin to a particular destination. This data is given from Mekelle city administrative office. Travel time is the time taken to reach a certain destination. Travel time estimation is applicable in different ways such as in determining traffic conditions in an area, the efficiency of the transport system, etc(Dimitris Sermpis, Charilaos Babis and Ioannis Theofilis). Travel time estimation is also important in modeling optimal public transport networks in congested urban networks, which can provide equitable service, because the travel time of public transport varies due to change in traffic volume (Krystian Birr, Kazimierz Jamroz and Wojciech Kustra, May 2012). Therefore, the estimated travel time also used in redefining or modeling the routes. The route length and the travel time for each route used in this thesis are shown in the appendix.

2.3.3. Productivity

According to Sami Jarboui, Pascal Forget and Younes Boujelbene, (2012)Productivity evaluates the output of an organization in relation to input in producing a product. Vehicle kilometers or daily passengers per vehicle measure the output of public transport. The inputs are the number of workers in the system, vehicles, and network length.

Vehicle kilometer: total distance traveled by a vehicle in operation per day. It is usually expressed in average kilometer per operating vehicle per day. It is influenced by traffic and road conditions, operating hours of operation, vehicle break down frequency and turnaround times. For a reasonably productive service, the vehicle kilometer is 210 – 260(Alan Armstrong -Wright and Sebastien Thiriez, 1987).

Passenger vehicle:is the number of passengers carried in relation to the capacity of the system. Passenger per vehicle can be measured in terms of the average number of passengers per operating vehicle per day(Alan Armstrong -Wright and Sebastien Thiriez, 1987). For buss with 80 numbers of seats, the passenger per day has to be 1000 – 1200. For bus with crush capacity of 100, the passenger per day is 1200 – 1500. For bus with crush capacity of 120, passengers per day is 1500 – 1800 and for bus with crash capacity of 160, passengers per day is 2000 – 2400 (Alan Armstrong -Wright and Sebastien Thiriez, 1987).Which is 12.5 – 15 passengers per seat so for minibus taxi with the crash capacity of 12 is 150 – 180 passengers per taxi per day.

2.3.4. Proximity

Proximity is the ease at which passengers are able to find a service. It usually indicated by the walking distance or area coverage of the service. The standard walking distance to public transport from home or work is 300 to 500m(Alan Armstrong -Wright and Sebastien Thiriez, 1987).Proximity measures close of service to origin or destination of people. By taking 500m as a standard walking distance by assuming people walk at 4 Km/hr. Proximity can be measured by simple GIS buffering (NIGER, 2011).

The stations have to be spaced at 500m distance and it has to be permanent, weather protected, with pedestrian information and has to have access to bike and pedestrian(NIGER, 2011).

2.3.5. Customer satisfaction

The customer is the ultimate judge of the quality of service. Effectiveness of minibus public transport identified through the customer’s perception survey. According to the same reference customer satisfaction, determinants for minibus public transport are classified into three(Harifah Mohd Noor, Na’asah Nasrudin,and Jurry Foo, 2014).

1 Comfort: Comfort is the main determinant for customer’s perception towards minibus public transport and transit bus public transport. Comfort includes convenience, physical condition, and comfort that a user experiences while he/she is inside the taxi or at a terminal waiting for transport services.

2 Accessibility: Is“the simplicity with which activities in the society can be reached, including needs of citizens, trade and industries and public services“NationalRoad Administration, (1998) sited in Maria-Christina Makrí and Carolin Folkesson.

3Safety: Safety shows use the safety feeling of safety towards using taxi transport during day and night. Respondent’s perspective in using taxi during day and night was positive so safety is not a problem in using taxis during the night but safety in taxi stations is more conservative.

2.4. Summary

In this chapter, efficiency indicators of public transport based on the literature review has identified. Those are infrastructure availability, quality, productivity, proximity, and customer satisfaction. The identified indicators further discussed in chapter 5 to meet the objective of the research. Lack of policy goals to measure the efficiency of public transport on the road and transport office makes efficiency evaluation difficult. The reference value for well-served areas in a developing country has taken from AlanArmstrong –Wright, SebastienThiriez (1987).

CHAPTER 3 METHODOLOGY AND MATERIAL

3.1 Introduction

This chapter deals with the data sources, sampling methods, data collection methods, analysis and interpretation techniques. In addition, it summarizes thespecific objectives, methodology, data required, and data source in a table.

3.2 Data source

This thesis has used both primary and secondary sources.

3.2.1Secondary data sources

Documented data, which relate to the population, trip production percentage of sub-cities, number of existing taxi transport vehicles and routes, were the secondary data sources. The major sources of materials were individuals and transport officials. Published materials on the structural plan of the city, planning and implementation strategies, maps in AutoCAD and ArcGIS format were also provided from the government institutions/ municipality.

Documents collected in relation to the objective of investigating the existing taxi transport problems were: on population growth, on spatial growth, on-road network, available road length and the type and so on. Studies on existing taxi transport routes were the documents collected from the city municipality.

Table 6:Summery of data collected from municipality

illustration not visible in this excerpt

3.2.2 Primary data sources

User’s perception on the existing taxi transport problems and on the factors to be considered in future design were the primary data sources. Data on the average trip per day, average network speed, average vehicle Kilometer and average passenger transported were primary data collected in the field.

3.3 Sampling

To determine the sample size of a study the population size, the level of precision, the level of confidence, and the degree of variability of the population are required. The sample size can be determined using a Census for small populations, using published tables, and using formulas (Israel, 1992). The sample calculation was done based on the simple formula byYemane (1967) sited in (Israel, 1992).

illustration not visible in this excerpt

Where

- n is sample size
- N is the population
- e is confidence level

The sample calculation for questioner respondent and taxi were done based on the above formula by taking the confidence level of 95%.

3.3.1. Sampling of Respondents

According to Mekelle city administration, population of Mekelle city in 2019 is estimated to be 360,475 as shown in appendix. From that population, sample size was calculated using the above formula to assess customer satisfaction using questioner.

illustration not visible in this excerpt

Therefore, samples of 400 customers weretaken to assess customer satisfaction. The sample distributions among each sub cities were based on the percentage of trip production by each sub-city as shown in the table below. The samples were taken randomly from each sub-city according to the sample proportion for each sub-city calculated using the trip production percentage of each sub-city. The thesis questioner was collected from different age, gender, income and occupation groups who use different taxi transport routes, of which 53% were male and 47% were female respondents. Samples were aged 10-70. Their income level was 31% below 500, 21% were between 500 and 1000, 15% were between1000 to 2000, 9% between2000 to 3000 and 24% of them were above 3000. In addition, work condition of respondents was 42% of students, 34% of government employees, 20% private employees and 4% were unemployed.

Table 7: sample distribution of respondents for each sub-city

illustration not visible in this excerpt

3.3.2. Taxi sampling

According to the data collected on field there are1050 taxi and 48 taxi routes in Mekelle city. Those1050 taxis in the city are distributed among the 48 routes. To find the number of passengers transported in each route, all of these taxis are used but in the evaluation of the average speed, the sample size was calculated using formula as shown below.The sampling method was random sampling.

In collecting the data on average loading, average vehicle Km and travel time of the taxis the sample size taken was

illustration not visible in this excerpt

Therefore295 samples weretaken and the taxis are selected randomly and distributed to each sub-city according to their trip distribution percentage as shown in the table below.

Table 8: vehicle sample distribution of each sub city

illustration not visible in this excerpt

“While travel time data can be collected for any period of the day, depending on the intended use of the data, the periods of greatest interest are usually those during which congestion occurs.” (TRAVEL TIME DATA COLLECTION, November 2008 ). In this thesis,the travel time and passenger transported per trip is collected at a time from 7:00 to 9:00 AM, 9:30 AM to 11:30 AM, and 6:00 to 8:00 PM to best represent the AM peak hours, free-flow hours, and PM peak hours, respectively. The data collection sheet is shown in the appendix part. The collected travel time is used to calculate the speed of the vehicles and evaluate the quality of the service by comparing it with the standards given by AlanArmstrong.

Whereas the vehicle Km is collected from,6AM to 9AM. The road and transport officials standing on each station, commonly called “teraaskebari” in Amharic, assigned to control the taxis in service collected this data.

3.4. Data collection techniques

The data collection for investigating the existing taxi transport problem and future suggestion were prepared in three types:

Type1. Questioner

This type of data collection method was used for customer satisfaction assessment.

The questioner used in this thesis has 4 parts as shown in the appendix part.

- Part I Background Information and Socio-Economic Status
- Part II Travel behavior of Subject
- Part III Feedback on current status of the transport system and future suggestions on modification.
- Part IVLiker-scale rating on customer satisfaction

Liker-scale was the fourth section of the questioner, whichwas conducted with the user respondents. In this section, respondents were asked to rate the statements from 1 up to 5 based on their level of satisfaction. The Liker-scale shows the degree to which respondents agree or disagree with the statements on customer satisfaction with a specified scale. Every respondent from the users was made to respond to all parts.

Type2. Field data collection

The researcher and data collectors using the data collection sheet shown in the appendix conducted this. It was an investigation on the direct parameters, which can affect the quality and productivity of the taxi transport system. The parameters were average loading, average travel time, and average vehicle Km. In collecting the data on these parameters data collectors were traveling in the taxis taken as samples for data collection and they were required to record the data.

illustration not visible in this excerpt

For the number of passengers, data were collected including those who do not travel end to end. After collecting the number of passengers transported in each vehicle, the average passenger transported per day found.

For travel, time the starting time, ending time and time lost in loading and unloading were recorded directly by the data collectors’ being in taxi and travel time is found by subtracting time lost and starting time from ending time. Then by dividing the route length to the travel time vehicle speed is found.

illustration not visible in this excerpt

For vehicle kilometer, the data collectors were standing at the taxi stations and counting the number of the cycle made by individual taxis per day and multiplying the number of the cycle by the length of the route resulted in vehicle kilometer.

illustration not visible in this excerpt

Type 3 Site observation

This part conducted by the researcher’s direct site visit on each route. It was an investigation on the condition of the route and vehicle condition. This helped the researcher to draw a conclusion on why the data has resulted in the numbers.

To sum up all,

1. Customer satisfaction was assessed through the data collected by questioner survey developed on factors that affect customer satisfaction of minibus taxi users.
2. Productivity and quality of the service was found from direct field data collection on average passenger per vehicle, average network speed and average vehicle kilometer. Then the values found from the analysis were compared with the value of each parameter for well performing minibus taxi transport system found from literature review to evaluate performance of the system.
3. Proximity: as minibus taxi transport are allowed to stop at any point first the current taxi routes were drawn in Arch GIS and then Arch GIS buffering were done at 500m.
4. Infrastructure availability: number of available minibus taxi vehicles were recorded on field and the route were drawn based on site observation. For this parametertransport modeling was used as analysis method having trip production and attraction for 2019 G.C from secondary data as an input then percentage of the available service was calculated dividing the total trip served by total trip demanded then the percentage coverage of the service was found.

3.5. Data analysis and interpretation

The methodology structured to address the research questions. To do this it is required to mix both qualitative and quantitative data, so a mixed-method research approach adopted in this study. This research approach involves mixing or combining quantitative and qualitative research techniques, methods, approaches, concepts or language into a single study(Johnson and Onwuegbuzie, 2004).

Mixing of both data types used fully to interpret the results found from the qualitative data type with the quantitative data type. In this thesis, the qualitative data will be the user's view in current taxi transport mobility (to investigate current situation), what do they wish to be modified, and this is done by asking the passengers using the questioner.And the quantitative data is the travel time data that will be collected in the field. Travel time is the time taken to reach a certain destination.

The thesis has used a mixed approach after the data collection. Then, data categorization was the first step. Data from different routes with similar types has categorized under similar categories. The data from site observation, field data collection, questioner and secondary data from the municipality on the existing number of vehicles, location, length and number of routes were used to assess the existing infrastructure availability, proximity, quality, customer satisfaction, and productivity. These data analyzed and used to evaluate the efficiency of the current taxi transport system based on the standards by Alan Armstrong. Whereas the data categorized under future recommendations on modifying the taxi transport system, the data collected by the questionnaire part 3, Feedback about the status of the current transport system and future suggestions on modification separated into qualitative information for proposing a new public route and taxi station.

3.6. Presentation technique

The thesis has used software like Microsoft excel and sigma-plot in the analysis part, MS Word and MS PowerPoint in the editing of simple graphics. Google earth used to visualize the city map. Arch GISalso used to draw the old and proposed taxi route and buffer the area under 500m walking distance from the city map, which found from the municipality.

3.7 Summery

The summary of the specific objective, data required, source of data and methodology used are summarized in the table below.

Table 9: Methodological framework

illustration not visible in this excerpt

CHAPTER 4 RESULT AND DISCUSSION

4.1. Introduction

This chapter deals with the results of the analysis based on the data collected and discusses the performance of the current taxi in the city by comparing the results with the value of the standards for a well-performing system.

4.2. Service productivity

Productivity is the extent to which public transport vehicles are put to use.According toAlan Armstrong -Wright and Sebastien Thiriez, (1987) service productivity of public transport is measured using passenger per vehicle and vehicle kilometer.

4.2.1. Passenger per vehicle

Is the number of passengers carried in a given time span usually a day. It shows the number of passengers carried in relation to the capacity of the system.As shown in the figure below routes with higher numbers of passengers per day are relatively short in length and paved with asphalt. Whereas those with low passenger per vehicle are routes with poor road condition and routes which pass through congested areas such as market.

illustration not visible in this excerpt

Figure 4:passenger per vehicle of each route

After identifying the outliers nine data were discarded and the average passenger vehicle per day is ≈170 passengers per vehicle per day as shown in the above graph.

For the well-performingminibus taxi transport system, 150 up to 180 passengers served per dayAlan Armstrong -Wright and Sebastien Thiriez,( 1987). Therefore,the current taxi transport productivity is in the range for the well-performing system so, the system is in good condition when evaluated in passengers per vehicle.

[...]