1 Introduction

The total number of trips and travel demand in Dhaka megacity, the capital of Bangladesh, is excessively high due to its population growth, urbanization rate, primacy, and economic activity [1, 2]. The existing transport system of the city is mainly road-based with a mix of different types of public transport, private personalized transport, and paratransit. According to the “Updating the Revised Strategic Transport Plan for Dhaka: Intertim Report-2 (URSTP)," traffic demand has experienced significant growth since 2014 to 2023 with around 37% increase in total number of trips [2]. However, trips carried by public transport, mainly bus, has fallen 50% since 2014 (from 5 million trips in 2014 to 2.5 million trips in 2023) due to poor service quality, inefficiency, safety issues, uncomfortability, and unreliability of this service. However, according to URSTP, the modal share of private vehicles has rapidly increased from 2014 to 2023: from 2.5 to 15.3% in case of motorcycle and from 18 to 24% for private cars. Alongside, the motorized three-wheelers, non-motorized rickshaws, walking, and bicycles have shown an increasing trend in the same period. But due to lack of regulation over their service quality, number, and operations have made them inefficient. Overall, URSTP outlines an obvious modal shift from public buses to private modes, mainly motorcycle. In such a context, there is a requirement for changes in transport system in the city and mode choice and trip pattern of the citizens.

It is found that throughout the world, MRT is argued, at least by government and authorities, as a game-changer in people’s mode choices and in existing modal split [3]. The introduction of MRT has the potential to contribute to the change of the existing modal split. Thus, many developing countries, especially in the context of Asia, are prioritizing MRT more with an expectation of a modal shift [3, 4]. Dhaka, the capital of Bangladesh, like many other fast-growing megacities, has planned to introduce five MRT systems with an expectation of visible modal shift and to address the growing transport demand. In consideration to the city’s transportation system, this shift could be from different types of transportation services like private personalized transport, public transit, and paratransit. Among the several proposed MRT Lines, MRT Line 6 was initiated at the first stage by constructing 20.1 km long route with sixteen (16) stations [5]. Initially, MRT Line 6 had been partially inaugurated in between nine (09) stations. Therefore, the introduction of a completely new public mode in this city of congestion has created a vibe among all the mode users, which has the potential to change the existing image of nearly incapacitated traffic scenario of the city.

In this context, studying the willingness to shift of the potential users is necessary to inform the decision-makers about the users’ perception and attitude toward a new public transport system. Studies in other context have captured the willingness to shift to MRT of mainly public transit and private vehicle users. Researches have also explored the factors behind modal shift and found context specific varied factors as discussed in the later section. Therefore, the firstly initiated MRT Line in Dhaka city has raised several research questions which are–

– What is the willingness to shift to MRT of different mode users in Dhaka city?

– How the people’s acceptance of MRT is related with their affordability, availability, and accessibility of MRT service?

– What are the factors influencing modal shift choice in the context of Dhaka city?

However, no research yet has captured the prelaunch willingness of potential users to shift to MRT and answered the raised research question in the context of Dhaka city. This study intends to address these questions and contribute some new knowledges in the existing literatures. The city has an exclusive variety in the mix of modes including around eight types of modes which can be broadly classified under public transit and different types of private personalized vehicle as well as paratransit users. The study on willingness to shift from other country contexts could only capture the scenario of different public transit (mainly bus and rail) and private transport users (mainly private car) depending on their context [4, 6,7,8,9,10,11,12]. Also, the literatures lack in assessing accessibility-related factors behind modal choice as discussed in later section. Additionally, a comprehensive approach is also missing while exploring the various influencing factors behind modal shift. Therefore, studying the willingness to shift to MRT and comprehensive assessment of the modal shift choice factors including demographic, socioeconomic, trip characteristics, and accessibility in the context of a unique and complex mix of modes structures has the potential to contribute new knowledge to the existing research base.

Therefore, the current study intends to understand the potential and willingness of different mode users to shift to MRT and the factors behind their choices. Findings regarding the people’s willingness to shift to MRT will inform the metro stakeholders about their perceptions, attitudes, demands, and characteristics as well as guide the relevant authorities in formulating strategies to hold on to the userbase and attract more. Also, the city has only operationalized MRT Line 6 and has plans to introduce five more MRT Lines in the upcoming days. Therefore, the study findings can guide the decision-makers regarding future development strategies that could facilitate shift to MRT.

2 Literature Review

Different countries around the world have introduced MRT to improve their existing traffic condition. Peoples’ travel behavior and modal shift potential have been studied during the introduction of MRT systems on a microscale in different country context. In a pre-feasibility study of MRT in Jakarta, a direct respondent survey was conducted to understand respondents’ willingness to modal shift which was analyzed based on their affordability, availability, accessibility, and acceptability [12]. Another study on the modal shift potential to the proposed MRT in the case of New Baneshwor and Radhe Radhe stations of Kathmundu revealed that around 66% of the different mode users are intended to shift to MRT [13]. The result of the revealed preference survey in one of the major cities of India, Mumbai, revealed that public transport users are the major shifters in the MRT [8]. However, recent studies in India found that development of mass transit are more feasible and now-a-days, private vehicle users along with intermediate mode users are being willing to shift to mass transit [14, 15]. While in case of Hanoi city, Vietnam, private car and motorcycle users are still reluctant to shift to urban mass transit instead of continuous improvements in public transportation system [4]. In the context of the UK, around 62% of private personalized vehicle users realize and feel encouraged to change their travel behavior for environmental goods, though only 42% of them are willing to shift to MRT from the use of private personalized vehicles [9, 16]. In Eastern European countries, private motorists are reluctant to shift to public transport even when those are made fare-free [11, 17].

Studies have viewed mode choice as a function of many factors including personal choice, socioeconomic, trip characteristics, spatial determinants, etc. [18]. In Bangkok, Thailand, a study employed revealed preference to compare the determinants behind modal choice between private vehicles and mass transit [19]. The model identified different socioeconomic and trip-related factors like gender, income, age, auto ownership, travel cost, trip time, income level, travel time, distance from the station, etc., as the significant determinants of mode choice. Similar method was used in a pre-feasibility study on the Riyadh metro of Saudi Arabia, where travel attributes like travel time, fuel cost, metro fare, and walking time showed significant impacts on modal shift to MRT of public bus and private car users though socioeconomic factors had marginal impact [10]. However, in China and Poland, a similar study identified mainly socioeconomic attributes having impact on modal choice where vehicle ownership has the most significant impact on modal choice between private vehicle and mass transit [18, 20]. In case of Mumbai, a revealed and stated preference survey method was utilize to model the modal shift behavior of public transport and private vehicle users where significant impact of different trip-related factors like waiting time, travel time, travel cost, and comfort level on the modal shift choice of the studied groups was found [8]. Using the stated preference method, the study of Hanoi city shows that the MRT could not attract many private vehicle users making long-term trips, but they are interested in shifting with necessary interventions [4]. While the case of Klang Valley MRT in Malaysia identified convenience, cost and time as influencing factors behind the modal shift from private vehicle users to MRT [21]. A study in Spain employed geographically weighted regression (GWR) model to focus on the spatial determinants of modal choice and identified active accessibility, respondents locations, bus lines as significant factors for the modal choice behind Madrid metro [22].

In summary, studies from different context have used mostly stated preference method in prelaunch study or revealed preference method in case of post-study or both for comparison. The studies have been mainly conducted based on either any specific public transit (bus or rail) or private vehicle users (focusing mainly car users and sometimes motorcyle users or both) depending on the specific context and research interest. Different factors have been found to have significant relationship with modal shift choice when analyzed using different logistic and spatial regression methods. Thus, the literatures have mainly studied context specific modes and factors while analyzing the willingness to shift to a new mass transit. Researches have not yet studied the scenario of willingness to shift in a context of varieties of modes including paratransit users and different private personalized vehicles users along with public transit users. Also, it is evident from the existing literatures that studies have widely explored different socioeconomic and trip-related factors behind modal choice while lagged in integrating various accessibility-related various factors like distance to the nearest station, time to reach the nearest station, influence of modes to access the station, etc. However, access to the stations of mass transit within reasonable time and convenient feeder modes is important [23]. Therefore, the factors behind the choice of mass transit including- demographic, socioeconomic, trip pattern, and accessibility are needed to be comprehensively evaluated.

3 Methodology

3.1 Research Framework

The overall framework of this research is presented in the Fig. 1. The framework portrays the entire research in summarized format. Firstly, research questions were raised based on the existing context and literatures were reviewed. This led to research gap identification and objective formulation of the research. After that, questionnaire was designed and survey method was decided. The research required primary data which were collected from the selected study area. Once data collection was completed, the data were analyzed to explore people’s MRT acceptance, willingness to shift of different mode users and factors behind modal shift choice. The analysis resulted in various findings which were interpreted and discussed in detail with respect to studies of other context, study context specific reasons and existing situation, key factors of modal shift choice, etc. With respect to the findings and discussion, policy interventions and different stick and carrot approaches by the authorities were suggested to attract potential users and hold on to the willing shifters to MRT.

Fig. 1
figure 1

Research framework of the study

Full size image

3.2 Selection of the Study Area

The study is based on MRT line 6, the firstly introduced metro service in Dhaka and in fact, in the country. The MRT line 6 is fully elevated with a total of sixteen (16) stations, of which initially, nine (09) stations came into operation. Out of those nine stations, six (06) stations and the interim links from the Dhaka city area were selected as the study area which includes Pallabi, Mirpur 11, Mirpur 10, Shewrapara, Kazipara, and Agargaon stations (fig. 2).

Fig. 2
figure 2

Map of the study route and the stations of MRT Line 6 Source: Adapted from [26]

Full size image

Those six stations were chosen for studying because they were going to be operational at the initial stage and were located within the city corporation area. The three (03) other stations were out of the city corporation area. The initial operational length was 12.1 km, and the study length covers 53% of the operational length [5]. However, it is to be noted that recently, the rest of the stations were opened. Thus, considering those stations, this study covered six (06) out of sixteen (16) operational stations. All the studied stations are situated in Dhaka North City Corporation (DNCC) area. Road alongside this MRT route is a two-way lane that supports mix of modes including bus, private car, motorcycle, rickshaw, bicycle, pedestrian, etc.

3.3 Questionnaire Design

The study followed a quantitative approach and used questionnaire survey method to collect the data. The study mainly focuses on the willingness to shift to MRT of the different mode users commuting in the roads aligning to the mentioned stations (in between Pallabi to Agargaon stations). The questionnaire included questions regarding their existing trip characteristics, demographic and socioeconomic status, their perception regarding MRT, expectations from the MRT regarding improvement of traffic scenarios, fare, headway, accessibility, and people’s acceptability, etc. (Fig. A-1).

3.4 Data Collection through Field Survey

The data were collected through a day-long direct questionnaire survey of the different mode users on 14 August 2022, four (04) months before the partial opening of MRT Line 6. The survey was conducted on a typical work day, from the morning to the evening peak, to get the best possible scenario and ensure representativeness. Eight types of mode users were reached while surveying. The respondents were those who travel in the road aligning to the studied MRT stations. The respondents were approached by the surveyors at the stations and in between in the interim links of the stations. The surveyors could easily reach the pedestrians but had taken the help of traffic police while surveying others. A stratified sampling technique was followed to reach the different mode users and then respondents were chosen randomly from each stratum. This method of sampling technique was used as there were different groups of mode users. A total of 410 samples, including public (bus), private personalized (car, motorcycle, bicycle, and walking), and paratransit (rickshaws, four-stroke motorized three-wheelers known as ‘CNG’, and four-stroke four-wheelers carrying 10-15 passengers, known as ‘leguna’) users were surveyed. The sample size is good enough considering 95% confidence level. Overall, the distribution of samples was 36.6%, 41.5%, and 22% among public transit users, paratransit users, and private personalized transport users, respectively. According to URSTP of Dhaka city, the modal share among the different passenger modes including: public transit, paratransit, and private vehicle users are 29%, 25%, and 46%, respectively [2]. This modal share calculation excludes pedestrians. However, walking being most common mode in Dhaka city and prominent in our study area was included within the definition of private personalized vehicle users. Therefore, the sample shares are slightly unmatched with the estimation of URSTP. In view of that, the distribution of sample among the strata is considerably representative of the existing modal share of different passenger modes in comparison to the URSTP’s estimation and the traffic scenario of the study area.

3.5 Data Analysis

Initially, the different mode users studied in this research have been analyzed using descriptive statistics. The potential shifter’s, which includes all the respondents, expectations toward MRT service have been comprehended based on their current monthly transport cost (affordability), expected headway of MRT service (availability), and preference for feeder modes (accessibility). Cross-tabulations have been performed to understand any significant variation in these expectations among the users’ groups and different demographic and socioeconomic groups. The willingness to shift has been analyzed based on potential users’ acceptability of MRT service and their conditions applied while shifting. The impact of influencing factors behind modal shift choice was analyzed by developing a binary logistic regression model. The results of the analysis have been portrayed using different types of graphs, charts, and tables and comprehensively discussed. Finally, policy recommendations are drawn compatible to the study findings.

4 Analysis of the Results

4.1 Profile of the Respondents

The sample has 41.5%, 36.6%, and 22% paratransit users, public transit users, and private personalized transport users including individual mode users and pedestrians. The demographic and socioeconomic profile of the respondents are summarized in the following table 1.

Table 1 Demographic and socioeconomic profile of the respondents
Full size table

Analysis of the respondent’s profile reveals that majority of the sample was aged between 18 to 30 years (44.3%) and 31 to 40 years (25%). The majority of the respondents (43%) had an income level between BDT 10,000 to 40,000 with nearly equal share among the income groups of BDT 10,000 to 20,000 (14.9%), BDT 20,000 to 30,000 (14.4%), and BDT 30,000 to 40,000 (13.7%). Most of the respondents were in service which could be government job or private job or any other type of job. Service holders are mainly distributed among three of these income groups (more than 50% of these three income groups) (Table B-1). Least of the respondents are no job holder which mainly includes unemployed or housewife or retired person. Around 21.5% had no income which was not due to unemployment. Rather, a significant share (71.9%) of respondents who had no personal income comprised of students (Table B-1). It is to be noted that few no job holders who had income level other than no income mainly includes retired person getting pension or other allowance. Around 17.5% had an income of more than BDT 50,000 and almost all of them were private car users.

The expectation of a heterogeneous road user groups from a completely new public transit with respect to affordability, availability of the MRT service, and accessibility by the integrated feeder services and their willingness to shift is analyzed below.

4.2 Affordability

The existing monthly transportation cost of different transport mode user groups (private personalized, public, paratransit) varies significantly and eventually impacts their cost expectation to shift to MRT. Among the private personalized transport users, private car users come from better a socioeconomic background than others, and thus affordability doesn’t seem to be a major issue for them. All the groups expect that MRT will be affordable for them, but there is variation in expected cost. Figure 3 depicts that different transport mode users' current spending on transport is quite dissimilar. As shown in Fig. 3, a portion of private personalized transport users spent mostly< BDT 1,000 monthly as transport cost, and they are pedestrians. On the other hand, the same group contains private car users whose cost is in the highest range.

Fig. 3
figure 3

Monthly Transportation Cost of Potential Modal Shifters to MRT Source: Field Survey, 2022

Full size image

Public and paratransit users’ current transport cost is mostly within BDT 5,000 and they expect to spend the same amount as bus fare when they shift to MRT (Fig 3). 26.3% of the potential shifters (considering all mode users) expect the same cost of MRT as the bus, while only 7% think that it will be doubled. These justify that reasonable fare structure is an important condition to shift a load from road transport to MRT. Almost 26% of the potential shifters from private cars have no idea about the cost, but they still think it will be affordable for them. Their monthly transportation cost is already too high (Fig. 3), and thereby, they perceive MRT to be less expensive anyhow. The expected cost scenario has no significant variation among different user groups. MRT needs to be affordable for all, irrespective of their current income, age, occupation, or transport mode.

4.3 Availability

Availability indicates if there are MRT stations near the respondents’ travel origin or destination at the desired time. Most of the respondents identified the unavailability of MRT station at their origin or destination to be the reason for not accepting MRT. Also, availing MRT depends waiting time at the station. In particular, private personalized transport and paratransit users can avail of their preferred transport whenever and wherever they need. About 80% of private personalized transport users and 44% of paratransit users’ current mode is ‘instantly available.’ In the case of public transport users, mode availing time varies from instantly available to 20 minutes. This largely depends on the traffic conditions and time of the day. Only 2.4% of all mode users said that they need to wait more than 20 minutes to avail transport now, and that is not even a regular case.

Around 55% of the potential shifters expect the similar availability of MRT as their current transport modes offer. There is noticeable variation regarding the expectation on availability of MRT among private personalized, public, and paratransit users (Fig.4). Around 88% of the respondents expect MRT to be available at the station within 5 minutes to 20 minutes.

Fig. 4
figure 4

Expected Headway of Potential MRT Shifters from Private Personalized, Public, and Paratransit Users Source: Field Survey, 2022

Full size image

Most of the potential shifters think that MRT will improve traffic conditions and no congestion will be faced while using MRT. That’s why waiting for 5 to 20 minutes will not be an obstacle in availing MRT service (Fig 4). Some of the private personalized transport users’ expectation of ‘instant availability’ of MRT questions whether MRT could be a reliable option for them or not.

Current headway in partially operating MRT is 8 minutes at peak hours and 10-12 minutes at off-peak hours [24]. But it is clear from the survey that people are expecting very frequent service of this mode, preferably within 5 minutes or 5 to 10 minutes is the expected headway by maximum respondents irrespective of the time of the day. However, the authority expects 3 minutes 30 seconds headway of MRT Line 6 after its full operation [25]. So, it is clear from the study that to catch these groups of people who expressed their willingness to shift to MRT, the expected headway of MRT must be ensured.

4.4 Accessibility

Accessibility includes the measures needed to avail of the MRT service. It is closely linked to availability. Nearly 70% of the potential shifters do not have any MRT station available at walking distance (considering 0.5 km to be walking distance). People will have to spend a significant amount of time to reach their nearest MRT station.

Of all the potential shifters, 35.4 % use rickshaws, 8% use leguna, 6.1 % use buses, 3.2 % use motorcycles, 2.7 % use bicycles, and 2.4 % use private cars to reach the nearest MRT stations. When potential shifters are divided as private personalized transport users, the dependence on walking and rickshaws to reach the nearest station is evident (Fig. 5). Thus, the operation of the MRT can increase the use of rickshaws as a feeder mode, and there might be congestion of rickshaws and pedestrians at MRT stations.

Fig. 5
figure 5

Mode Available to Potential Shifters to MRT to Reach the Nearest Station Source: Field Survey, 2022

Full size image

Figure 5 shows that a portion of existing private car users would like to use the car as feeder mode. During survey, it was found that on average they need to travel 1.3 km to reach their nearest station, which is longer than the average first/last distance of rickshaw or leguna users. These car users expressed their willingness to use the MRT provided that car parking facility is provided for them in the MRT station areas. Leguna, which is a paratransit, seems to be less used. Reasons may be lack of availability of the service, poor quality of service, or lack of information regarding its routes. However, the mode has the potential to contribute to increasing MRT accessibility.

The variation of feeder mode choice among different groups is noteworthy (Fig.5). Despite this situation, only 21.8% of the potential shifters stated that the introduction of a new mode can enhance accessibility to MRT stations. Though walking is preferred mode to access MRT, only 33% of potential shifters demanded the improvement of the walking environment. There are scopes to increase the accessibility to MRT by providing transports as well as other infrastructure facilities.

4.5 Willingness to Shift

After having discussed of affordability, availability, and accessibility of the potential shifters regarding the upcoming MRT service, their acceptability or willingness to shift to this new mode is about to be discussed now. The results regarding willingness to shift to MRT and the type of conditions stated by them are revealed in Fig. 6.

Fig. 6
figure 6

Willingness to shift by different mode users Source: Field Survey, 2022

Full size image

On the whole, almost 75% or more users of this route are willing to shift to MRT. Mode-wise data shows that 96% of paratransit users are willing to shift. If paratransit users are further categorized, it is found that around 98% of the rikshaw users are willing to shift to MRT. Private personalized transport users are comparatively less willing especially, pedestrians (72%), and motorcycle users (73% of them). However, 88% of bicycle users, a sub-category of private personalized transport users, would like to shift. Around 85% of the different mode users are willing to shift either unconditionally or conditionally with a share of around 34%-66% accordingly (Fig. 7). However, the conditions put are different by the mode users who have willingness to shift. Private personalized transport users have mostly prioritized comfort issues like air conditioning (20%), less crowding (18%), etc. Particularly, private car users are highly demanding of these comfort issues (i.e., air conditioning (33.4%), less crowding (33.2%)) which are compatible with their current mode behavior. Bicycle users have put conditions like frequent availability of MRT service (30%), and motorcycle users, have prioritized time-saving (18%). Public transit users are demanding less crowding (22%) and safety (18%) as conditions to shift to MRT. The overcrowding situation of buses and safety-related issues, like women's harassment, pickpocketing, etc., have made them demanding of these conditions. Fare of MRT is the big priority concern of 30% of paratransit users, especially leguna users (50%) while making a shift.

Fig. 7
figure 7

Conditional vs. unconditional shifting Source: Field Survey, 2022

Full size image

4.6 Factors Influencing Modal Shift

A binary logistic regression model has been developed in RStudio Software to identify the factors and their significant impact on modal shift choice. The choice to modal shift to MRT is a binary categorical variable (Yes/No) that can be dependent on different factors. Firstly, the model’s predictors were decided. The factors (independent variables) that were initially considered for the model include demographic: gender, age-group; socioeconomic: income, monthly transportation cost, vehicle ownership, educational status, and occupation; trip-related factors: trip origin and destination, trip purpose, trip time, trip frequency; and accessibility-related factors: nearest station’s distance, time to reach nearest station and modes to access the station. Some of the categorical variables were regrouped to maintain the balance in frequency among categories (if necessary) and then converted into dummy variables before entering into the model. After running some trial and error, the predictors that were found to be significant in all trials were considered for the final model development. The final model includes 10 predictors. Table 2 describes the independent variables. The developed model’s Wald test result yielded a significant model at 99% confidence level with Chi-square statistics of χ2(4) =74.7, p < 2.7e-10. This indicates that the model is a significantly better fit to the data that explains the relationship between the choice of modal shift and different influencing factors.

Table 2 Description of the model’s independent variables
Full size table

After model development, it has been validated by checking the assumptions. Table 3 describes the assumptions of a binary logistic regression model in case of the developed model.

Table 3 Model validation based on assumptions
Full size table

None of the assumptions of a binary logistic regression model are found to be violated in case of the developed model which make it a fit one. Table 4 shows the model’s results and the interpretation from the results is discussed in terms of the odd ratio below-

Table 4: Model summary
Full size table

4.6.1 Sociodemographic Factors

Gender: Male respondents are around 1.57 times or 157% (2.57002119-1= 1.57002119) more likely to shift to MRT than female respondents.

Vehicle ownership: Respondents who own any type of vehicle are around 18% (1-0.81467095=0.185329) less likely to shift to MRT than those who do not own any vehicle.

Educational status: Respondents who have SSC to HSC level and higher education are around 98% (1.98250260-1=0.98250260) and 439% or 4.39 times (5.38573016-1=4.38573016) respectively more likely to shift to MRT than those who have below SSC level education. Thus, the higher the people are educated the more they are positive about modal shift.

Occupation: Respondents in service are around 129% (2.28628449-1= 1.28628449) or 1.29 times more likely to shift to MRT than those with no job. Respondents doing business are around 85% (1- 0.14934651= 0.850653) less likely to shift to MRT than those with no job.

4.6.2 Trip-Related Factors

Origin or destination of the trip: Respondents who are originated or destined to locations that are completely non-aligning to the MRT route are 54% (1- 0.46315044= 0.53685) and 51% (1-0.48920972= 0.51079) less likely to shift to MRT than those whose origin and destinations of the trip considerably align to MRT route.

Trip time: As the trip time prolonges, respondents become less likely, around 10% (1- 0.89588899= 0.104111), to shift to MRT. Therefore, longtime trip makers are less willing to choose MRT.

Trip frequency: Respondents who make trip on regular basis are around 47% (1.47560929-1= 0.47560929) more willing to shift to MRT than those who make trips on irregular basis.

4.6.3 Factors Related to the Access to MRT Station

Time to reach the nearest MRT station: As the time to reach the nearest stations increases, respondents get less likely, around 31% (1-0.68789059= 0.312109), to shift to MRT.

Modes to access the MRT station: Respondents who have considered public bus or private vehicles or walking as a mode to access the MRT station from their available modes are around 72% (1-0.27840822= 0.721592), 95% (1- 0.05236843 = 0.947632), and 85% (1- 0.14596581= 0.854034) respectively less likely to shift than those who have considered paratransit to access MRT station. Thus, availability of paratransit as a feeder service to access the MRT station is an important factor to facilitate the shift to MRT.

5 Findings and Discussion

The findings of the study suggest the decision-makers to take prompt action to transform the expectations of the potential users into reality to hold on to the potential user base. The prelaunch study on MRT Line 6 finds a high share of willing shifters to MRT among all users. The willingness is more evident among the para and public transit users than private personalized transport users. However, one of the main goals of MRT introduction in Dhaka city was to reduce the dependencies on unsustainable mode use including cars, motorcycle, etc. private transport and reducing their excessive pressure on the major roads. The similar studies in the context of many Asian countries including: India, China, Malaysia etc., Saudia Arabia, Eastern America had found that public transit users are the most willing ones while making a shift [7, 8, 10, 11, 17]. But in our context, the transportation mode structure being complex and unique than others, reveals a different scenario and finds paratransit users to be the most willing shifter to MRT. As paratransit users’ shift to MRT was not studied in other research, their significant shift is a new finding and not comparable with other country scenarios. Among different paratransit users, rickshaw users are found to be highly willing to shift. This can be explained by the fact that the MRT is more frequent, much speedier as well as cost-saving compared to rickshaw, which is often the most suffering mode on the roads of Dhaka city in the midst of other motorized modes. However, similar to the other studies the willingness of public transit users o shift is also evident in our context which is explainable in consideration of the current status of this mode like poor quality, unsafe, overcrowding, and, most importantly, high immobility during traffic jam compared to other modes in the roads of Dhaka. Also, the conditions put by public transit users while making a shift are similar to the problems they face in their current mode. The study in China and Mumbai having similarities with our context, revealed the similar conditions of shifting where public transit users were more willing to shift than private vehicle users [7, 8]. Though the share of willing private personalized transport users is comparatively lower than other two types of mode users, it is still high when compared to other contexts. In fact, 75% of private motorists (car and motorcycle users) would like to shift which is much higher than the willingness of 60% of private motorists in European and Asian countries [4, 8, 16]. Perhaps, the introduction of a new mode, particularly a much talked, publicized, capital, and technology-intensive mass transit system for the first time in the entire country, might influence their willingness. Also, the recent price hike, close to double, of fuel might have an impact on their responses. However, most of the users are willing to shift conditionally and do not come out of their comfort zones associated with existing modes. Rather, the users of public and paratransit are demanding MRT service to overcome the problems associated with their current mode of use. Also, they want some more benefits which are not offered by their current modes. The current modes of private personalized transport users are already offering better services which might have influenced their willingness to shift. However, to make the expected modal shift, MRT service needs to fulfill these wide varieties of demands coming out of the potential shifters and offer some more benefits to attract more users.

The findings of the studied scenario of Dhaka match with a pre-feasibility study of MRT in Jakarta, where respondents’ willingness to modal shift was analyzed based on affordability, availability, and accessibility [12]. A positive relation between metro acceptance, reduced distance from the nearest metro station and lower fares of MRT is consistent with the findings from a study in Saudi Arabia [10]. The case of Dhaka city reveals that MRT being a mass transit, needs to ensure its affordability as people of the city expect it to be like a public bus. Introducing a variety of card or pass systems based on socioeconomic classes of people can ensure the affordability of MRT. Concession on MRT fares is common in the American as well as in Australian context including free riding for children and elderly people with age limits, reduced fare for mobility challenged person, discounts on fares specially for elderly, height limits for children who can enjoy free rides, low-income fare rides etc. On the other hand, practice of such fare incentives is less likely in Asian cities with some exceptions in Shanghai, Guangzhou, Kuala Lumpur, Tokyo etc. However, discounts on fares for students and special fare services for tourists are available in almost all metro systems worldwide. In Dhaka city, no fare incentive system is introduced yet for the MRT Line 6. The city’s metro has MRT pass or Rapid pass system that only ensures the time-saving by avoiding long queue. The availability of MRT service in the city needs to be improved and more instantaneous than its current frequency to match people’s expectations which is less than 5 minutes mostly. On the other hand, the current frequency is on average 10 minutes which is double to the potential user’s expectation. Also, modal integration, with feeder services needs to be ensured by the respective authority to improve its accessibility condition. In this regards, our neighboring country India has introduced feeder bus services integrated with Delhi metro [26]. But in Dhaka still there is no effective initiative for feeder services yet causing difficulties to MRT users while accessing stations.

While analyzing the factors influencing modal shift, the study found different sociodemographic, trip characteristics and accessibility-related factors having significant impact. Though the choice to modal shift according to different age-group, income, monthly transportation cost, trip purpose, distance to the nearest station is not significantly associated, different studies have found these factors having significant impact on modal shift choice based on their context [6,7,8, 10]. Our study found male users to be more willing to shift to MRT than women, though the recent intervention of women-only compartment designation by the authority is much appreciated and could attract them more. Regarding vehicle ownership, peoples who have personal vehicle are less willing to shift which is quite common. In this regard, many Asian countries like Taipei, South Korea, Japan, Hong Kong, Thailand and the Philippines took intervention to prohibit motorcycle ride on arterial and sub-arterial roads [4]. However, no strict or carrot approach have been initiated yet by the authority in Dhaka to attract them. The study also found that people belonging to lower-education status (mainly those who have education below secondary level) are comparatively less willing to shift. This mainly includes the students who are still pursuing education. Also, the no job holders (mainly including students) are less willing to shift. This can be explained by the fact that the authority has no plan yet to initiate student pass in metro which is a very common approach worldwide. The less interest to shift to MRT by people in business can be explained by the fact that, they might be in need of freight movement which is not supported by MRT system. Regarding trip characteristics, it is found that the people who have origin and destination of their trips aligning to MRT route are more willing to shift. This can be explained by the fact that routes that are non-aligning to MRT are not yet connected to the stations by feeder services or other facilities which might have discourage them. However, to attract more people to use this mass transit until the launch of other MRT routes, connectivity of the currently operated MRT line 6 with non-aligning routes is important to ensure. Regarding trip time, people making prolonged trips are less willing to shift. However previous literature found users are more likely to shift to MRT while making lengthy trips for saving cost, time, and more comfort [4]. This contradictory finding can be explained by the fact that only a segment of MRT line 6 is going to operate where the stations are closely situated [5]. Also, the fare is an important factor as longtime trip on MRT covering longer distance requires comparatively higher fare than public transport like bus. In addition, MRT being a completely new mode people are not habituated to it yet. Thus, making prolonged trips in a completely unknown service might not be comfortable for people at the initial stage. The result regarding trip frequency says that regular trip makers are more willing to shift which is also supporting to the result of service holders willingness to shift who are mainly involved in frequent trip making. Thus, MRT needs to be reliably operated on regular basis to support those frequent trip makers. The current operation of MRT is kept restricted on Friday for maintenance purpose. However, Friday being weekend in our context, such restrictions will not cause major issues. While analyzing the factors, this study took an attempt to identify the impact of accessibility-related factors on modal choice. It is found that prolonged time to reach the nearest MRT station have negative influence on shift to MRT which can be addressed by improving the accessibility through introducing efficient, fast and reliable feeder services. Regarding the access to stations by modes, the study found new and interesting finding. Paratransit as available modes to access the MRT station has profound impact on the choice of MRT than the availability of other modes. Therefore, our context specific different paratransit like rickshaws, leguna or even mini bus type modes (similar to leguna) can be introduced as feeder services in the internal roads connecting the MRT stations. This can facilitate attracting people whose trips are originated or destined to the routes non-aligning to MRT line 6 by ensuring access to the stations through reliable services in an efficient manner.

6 Policy Implications

The findings of this study underscore the importance of measures by decision-makers to align with the expectations of potential users. The study advocates the following recommendations based on the findings:

- Affordable fare policy: As people from different socioeconomic classes are expecting MRT’s fare to be affordable similar to public bus, it is necessary to introduce various fare options, including discounted fares for students, senior citizens, and low-income individuals, to ensure its affordability. However, determination of metro fare based on these criteria is already included in “Metro Rail Act 2015” which needs proper implementation [27].

- Ensuring frequent availability of MRT: The study finds people’s high expectation regarding MRT’s frequent availability, preferably less than 5 minutes. The authority is also committed to ensure 3 minutes 30 seconds headway [24]. This commitment is yet to be fulfilled, even since one-year of MRT launch, as the current frequency is between 8 to 12 minutes, Thus, the policy regarding MRT headway should be implemented forthwith.

- Gender sensitive measures: As the study shows difference in modal shift between male and female, gender-sensitive policies including designated waiting areas for women, enhanced security measures, and women-only compartments, to improve the safety and comfort of female passengers should be implemented. Dhaka Mass Transit Company Limited (DMTCL) has taken a “Gender Action Policy 2015’ which covers all these aspects and the study justifies the presence and implementation of this policy [28].

- Modal integration and feeder services: The study reveals that potential shifters demand establishment of efficient feeder services to access MRT stations. Unfortunately, the “National Integrated Multimodal Transport Policy 2013” (NIMTP 2013) does not state how MRT as an urban rail transit system will be integrated with other modes [29]. However, the study finds accessibility to the MRT stations by different paratransits as an important factor behind modal shift to MRT. But the integration of paratransit modes is not mentioned properly rather the policy guides on banning rickshaw gradually on feeder roads. Hence, the policy makers need to rethink their decision regarding paratransit modes and take policies to ensure modal integration by feeder services.

- Regulating personal vehicle use: The study finds comparatively less willingness of private vehicle users specifically, motorcycle users and vehicle ownership as a deterrence toward shift to MRT. This demands some strict or regulatory approaches from the authority. The “Urban Transport Policy” (UTP) 2015 has already outlined some of the regulatory measures including taxation, parking fees, yearly charging, area pricing, designation of “motorized-free zones” by times or days [30]. Unfortunately, such zones are not designated yet. Therefore, it is high time that the authority identify such zones which can be aligning to metro route to attract potential users.

- Parking and turning facilities for accessibility: The study suggest short time parking and turning facilities to be provided at the MRT stations for paratransit modes which are found to be the most prioratized feeder services. Also, the stations being already developed, the study suggests provision of underground parking and at-grade turning facilities for private vehicles to attract the private motorists more toward modal shift. The “Draft Parking Policy 2019” also guides to facilitate park and ride services at MRT stations which need to be implemented [31].

7 Conclusion

The study focuses on the modal shift of different mode users considering the imminent launch of MRT Line 6 in Dhaka city. The introduction of MRT involves modal split expectations. This pre-study reveals 75% or more of willing shifters, whether they are public transit or private personalized transport or paratransit users. Specifically, the paratransit users hold the highest share (96%). This study, compared to other studies from different contexts, has identified more private personalized transport users who are willing to shift. These findings provide a positive indication regarding the future modal split by different mode users. However, users would like to shift mainly based on one or more conditions. Public transit users demand less crowding and safety as they face the opposite in public buses, a big concern for women. Private personalized transport users are demanding comfort, availability, time-saving, etc., which are compatible with their current mode of use. Hence, MRT must uphold its image to be safe, comforting as well as a fast transit. The study also finds some sociodemographic, trip characteristics and accessibility-related determinants regarding the modal shift choice. Respondents being female, less educated, having no job or being business man, making irregular and longtime trips, originated or destined to areas non-aligning to MRT route, facing longer times to reach MRT station and having access modes other than paratransits are to be less likely to shift to MRT. Therefore, to hold on to the potential shifters and attract more of them, decision-makers need to play their role by taking necessary interventions, implementing the existing policies and applying relevant strict and carrot approaches. Fare incentives for different socioeconomic groups specially, student pass system, acheiving desired frequency, modal integration with efficient feeder services, prohibiting personal vehicle use (like cars, motorcycles) as well as parking facility for those, etc. could be initiated by the authority. The introduction of MRT has made maximum of the different mode users optimistic regarding the improvement of future traffic scenarios. Thus, the findings of this research inform the decision-makers about the expectations as well as demands of different mode users while making a shift to MRT which need to be confirmed by effective policy decisions. Considerations of users’ standpoint can ensure the expected modal shift to MRT and make it more viable.

This study being a prelaunch study requires comparison of its results with any postlaunch study in the context of Dhaka to assure its correctness. Therefore, there is further scope of revealed preference study and then, those findings can be compared. Since the route of MRT Line 6 is one of the most important transport corridors of Dhaka, there is already the presence of various transport modes. So, the competitiveness of MRT compared to other modes remains an intriguing question that calls for further study. Also, similar studies can be attempted before the launch of other MRT routes to better comprehend the users’ perspective and realize their expectations and demands.