Introduction
Rapid urbanization, population growth, and increased national productivity globally have exacerbated land transportation challenges in numerous cities, adversely impacting environmental, social, and economic conditions. The efficiency of a city's transportation system is pivotal to its economic and social well-being, shaping urban structure, influencing economic progress, and directly impacting residents' quality of life. Urban transportation planning, facilitated through the four-step transportation modeling system (UTMS), is instrumental in addressing these challenges by estimating transportation demand, evaluating the movement of people and goods, and optimizing transportation infrastructure and services within metropolitan areas.
The four-step transportation modeling process, a widely adopted approach in urban transportation planning, involves successive stages: trip generation, trip distribution, modal split, and trip assignment. These steps collectively assess the generation and distribution of trips, mode choices, and route assignments, respectively. Origin-destination surveys, particularly utilizing home interview methods, are integral for gathering data on the movement of persons and vehicles within an urban area.
This study seeks to provide comprehensive insights into public transportation challenges in Asian developing nations, with a specific emphasis on major area in Myanmar. Positioned as a significant trade and transportation hub between China and India, Myanmar faces unique difficulties due to its diverse topography, which includes mountain ranges, hills, and valleys, contributing to infrastructural limitations. The exploration of public transportation issues in Myanmar, exemplified by the examination of Mandalay, aims to enhance our understanding of the complexities inherent in developing effective transportation solutions in the region.
Study Area Profile
The 18 wards( the smallest electoral unit) of Mahar Aung Myay Township of Mandalay City Corporation area have been selected as the study area for this paper (Fig. 1). Then these 18 wards are divided into 5 zones known as TAZ (Traffic Analysis Zone).
Fig. 1 Network for Mahar Aung Myay Township
Methodology of the Research
The traditional four step transportation modeling system has been taken to achieve the objectives. This is a macro-level working procedure. The following four steps to performed in the next stage:
Trip generation
Trip generation is the first stage in the traditional four-step transportation planning method,which is extensively used for projecting travel demand.It forecasts the number of trips that originating in or destined for a particular traffic analysis zone.
Trip generation employs trip rates that are averages for a wide portion of that study area. Trip production are defined as the return end of a home-based trip or the starting point of a non-home based excursion.Some of the factors considered as predictor for trip production include family income, vehicle ownership, and the number of worker per home. For example, a household with four people and two vehicles may be assumed to produce 3.00 work trips per day. Trips per household are then expanded to trips per zone. Trip attractions are typically based on the level of employment in a zone. For example a zone could be assumed to attract 1.32 home based work trips for every person employed in that zone. Trip generation is used to calculate person trips.
This stage determines trip production and trip attraction after ten years (2023) (based on 2013). To begin, present trip production and attraction characteristics are estimated using 10-year growth rates (Table 1). These growth rates are based on a country-by-country aspect. Regression equation for production model is based on household characteristics of the study area. In this study, household characteristics used in production model are average income, average household size and average worker as shown in Table 2. Calculation of the attraction model is based on the zonal floor areas such as retail areas, non-retail areas and other areas of the study area. They are obtained from Mandalay City Development Committee (MCDC) and shown in Table 3.
Table 1. Growth rates of different variables after 10 years
Table 2. Zonal information of Mahar Aung Myay Township of Mandalay
Table 3. Retail, non-retail and other areas in Mahar Aung Myay Township
Regression model is useless in null hypothesis (H0). Under the null hypothesis, the ratio of the means of the two respective sums of squares is denoted F. The null hypothesis is rejected if F > Fp, (n-p-1) for the level of significance. The mathematical expression is
\(\begin{align}F=\frac{V_{R}}{V_{E}}\end{align}\) (1)
where; VR = the mean of the sum of square due to regression
VE = the mean of the sum of square residual
VR and VE can be calculated by the following equations.
VR = SSR/degree of freedom (2)
Trip distribution
Trip generation is the first stage in the traditional four-step transportation planning method, which is extensively used for projecting travel demand. It forecasts the number of trips that originating in or destined for a particular traffic analysis zone.
Travel times are in the form of a matrix; each cell represents the time it takes to travel from one zone to another zone. They are shown in Table 4 and Table 5.
Table 4. Zone to zone travel time
Table 5. Observed trip distribution matrix, Qij0
The travel time and cost for each mode are basic parameters of modal split model. Household characteristics, travel time and cost of the traveler are used independent variables to get the typical utility equation for each mode. In this study, household income is used as household characteristics.Mean speed is obtained from speed study and number of vehicle per hour and length are obtained from ground count, shown in Table 6.
Table 6. Number of vehicle per hour, Vf, mean speed and length for each link
Result
Modal split
Mode choice analysis is the third step in the conventional four-step transportation planning model. Trip distribution's zonal interchange analysis yields a set of origin destination tables which tells where the trips will be made; mode choice analysis allows the modeler to determine what mode of transport will be used.
In this study, utility function is used to calculate utility of each mode. Four modes are considered as car, motorbike, bicycle and bus. Three independent variables are considered in calculation of utility equation. They are income, travel time and travel cost. The dependent variables are considered by the number of utility of mode choosing of travelers.
Multinomial logit model
In light of its simplicity of estimate and grounding in utility theory, the multinomial logit (MNL) model is an extremely often used model to explain and forecast discrete decisions. The MNL model is a generalization of the binomial choice model to more than two possibilities. The multinomial logit model calculates the probability of choosing mode ‘K’ if disaggregate or the proportion of travelers in the aggregate case that will select a specific mode ‘K’ according to the relationship.
\(\begin{align}P(k)=\frac{e^{U_{k}}}{\sum_{x=1}^{n} e^{U_{x}}}\end{align}\) (3)
In this study, multinomial logit model is used to determine the probability of each mode by using Equation 3. Then, zone to zone O-D matrix by each mode is received by multiplying probability of each mode and zone to zone trip interchange volume obtained from distribution model. These results are shown in Table 7 to Table 10.
Table 7. Zone to zone O-D matrix by car
Table 8. Zone to zone O-D matrix by motorbike
Table 9. Zone to zone O-D matrix by bicycles
Table 10. Zone to zone O-D matrix by bus
According to the results of above tables, it is found that zone-5 has the maximum percentage of flow for each mode. Moreover, zone to zone trip distribution by motorbike is more than other modes. Therefore motorbike trips are the maximum among all zones.
Trip assignment
Trip assignment, traffic assignment or route choice concerns the selection of routes (alternative called paths) between origins and destinations in transportation networks. It is the fourth step in the conventional transportation planning model. Mode choice analysis tells which travelers will use which mode. To determine facility needs and costs and benefits, we need to know the number of travelers on each route and link of the network.
The task of the assignment process is to establish the loading, or user volume on each link of a transportation network. Therefore, the length of each link and the volume of traffic flows on that links are necessary for this assignment model. Also, the mean speed for specific route is needed. Free-flow speed (FFS) of each link is calculated by using Equation 4. The results of free-flow speed (FFS) for each link are shown in Table 11.
\(\begin{align}F F S=S_{F M}+0.0125 \frac{v_{f}}{f_{H V}}\end{align}\) (4)
Table 11. Free flow speed
In the calculation of free-flow speed (FFS), mean speed (SFM) is obtained from speed study and vf is the number of vehicle per hour. It is also known as the observed flow rate for the period and obtained from ground count. The flow of traffic with unrestricted mixing of different vehicle classes on the roadways forms the mixed traffic flow. Therefore, the observed flow rate, vf is converted to the homogeneous traffic consisting of passenger cars only by using Passenger Car Unit (PCU). The heavy vehicle adjustment factor of 0.829 is calculated by using Equation 5.
\(\begin{align}f_{H V}=\frac{41}{\left[1+P_{T}\left(E_{T}-1\right)+P_{R}\left(E_{R}-1\right)\right]}\end{align}\) (5)
Capacity is expressed as the maximum number of vehicles in a lane that can pass a given point in unit time, usually an hour, i.e., vehicle per hour. Capacity for each link can be calculated by using Equation 6. Effective green time to cycle length ratio, q is considered as 0.5. The results of capacity for each link are shown in Table 12.
C = 1900 × L × q (6)
Table 12. Assigned travel flow and travel time
Travel time, T can be calculated by using Equation 7.The results of capacity and travel flow of each link are used in the calculation of travel time. Free- flow travel time is the ratio of length to free-flow speed. The results of travel time are shown in Table 13.
T = T0[1 + 0.1519v/CP)4] (7)
Table 13. Travel time T(x0)
According to the Table 13, link (26-5) and link (29-27) are congested by comparing free-flow travel time. Travel flows of link (26-5) and link (29-27) are shifted by using Frank-Wolf algorithm. The travel flows on link (26-5) are shifted to link (25-26), (25-4) and (4-5). The travel flows on link (29-27) are shifted to link (29-32), (32-33), (33-34), (34-28) and (27-28).Table 14 shows the results of travel flow and travel time after shifting the links.
Table 14. Assigned travel flow and travel time
After shifting the links, travel time on link (26-5) is reduced from 25.92 minutes to 7.85 minutes and link (29-27) is reduced to 6.26 minutes to 3.2 minutes. Zone to zone travel times are obtained by combining link travel times. They are shown in Table 15.
Table 15. Zone to zone travel time
In Table 13, interzonal travel time of zone-3 and zone-5 has 17.1 minutes which is the longest travel time among the proposed networks of the study area. Because of that path contains two congested links, 29-27 and 26-5 which have the travel time of 3.2 and 7.85 minutes respectively.
Conclusion
The study area is divided into four wards, with a sample size of 400 established for the home interview survey. The generation model considers average income, household size, and vehicle ownership as independent factors, while the number of trips is the dependent variable. Coefficients for the trip generating model are calculated using the Linest formula, and the model is validated using R2 and F-test. Zone-5 produces the highest number of trips at 86863.
In the attraction model, retail locations, non-retail areas, and other places are used as independent variables. The regression equation is employed to compute attraction trips for each zone, and the model is validated using R2 and F-test. Zone 3 records the highest attraction trips at 71325. The distribution model involves four iterations to estimate the friction factor, Fij. The modal split model, utilizing the multinomial logit model, considers four forms of transportation: automobile, cycle, bicycle, and bus. Results show that motorbikes are the most popular means of transportation in the five zones.
The assignment model establishes linkages and nodes for the research area, utilizing trip interchange matrix data for each connection. Survey ground counts and traffic volume counts are considered in the analysis and design of traffic signals. The assignment model factors in the use of car, motorcycle, and bicycle, converting person trips to car trips and then to peak hour and Passenger Car Unit journeys. Travel times for specific links are assessed, revealing crowded connections. After adjusting traffic patterns, travel times for certain links are significantly reduced. Interzonal travel times, calculated by adding link travel time, show the longest trip time in the planned networks is 17.1 minutes from zone-3 to zone-5 due to crowded links on that path.
References
- Ahmed, B. (2012). "The traditional four steps transportation modeling using simplified transport network: A case study of Dhaka City, Bangladesh." International Journal of Advance Science,Engineering,Technology & Research, Vol. 1, No. 1, Article #03.
- Aminzadeh, B., Levinson, D. (2020). Transportation Impacts of Automated Vehicles. In Autonomous Vehicles and Future Mobility, Elsevier, Netherlands, pp. 101-119.
- Chakirov, A., Susilo, Y., Kottenhoff, K. (2020). "Intermodal connectivity and passenger travel behavior: A systematic review." Transport Reviews, Vol. 40, No. 2, pp. 235-262.
- Curtis, C. (2015). The Congestion Myth: Living with Traffic Jams. Reaktion Books, London, United Kingdom
- Hensher, D.A., Mulley, C., Ho, C. (2017). "What drives bus rapid transit system patronage? A comparative analysis of metropolitan cities with different levels of bus rapid transit system development." Transport Policy, Vol. 54, pp. 43-52.
- Mahmassani, H.S. (2018). "Smart mobility: Connecting everyone and everything." Transportation Research Part A: Policy and Practice, Vol. 115, pp. 4-5.
- Zhang, J., Gao, Z., Wang, D.Z.W., Chen, Z. (2017). "Traffic congestion management in smart cities: Challenges, methodologies, and opportunities." IEEE Communications Magazine, Vol. 55, No. 1, pp. 106-112.