Leveraging Battery Electric Vehicles (BEV) and Policy Interventions to Mitigate Jakarta's Traffic and Pollution Crisis: A System Dynamic Analysis

3.1 Constructing of system dynamic model

The initial steps of the System Dynamics (SD) process involve defining the modeling goals and identifying relevant system components. The first task is to gather data that relates to the problem at hand. After data collection, each variable must be analyzed to understand the correlations among them. This paper focuses on three key areas to address the problem: clustering roads to assess their impact when the mandatory BEV policy is implemented; determining the scale of BEVs and ICE vehicles to identify when the BEV population exceeds that of ICE vehicles; and evaluating the effect of reduced ICE numbers on air pollution and future traffic congestion influenced by the growth of BEVs.

3.png

Figure 3. Main flow chart for system dynamic

Once the problem analysis is completed, the next step involves creating a simulation, beginning with the construction of causal loops and flow charts that illustrate the relationships among the factors identified. The resulting stock-flow diagram consists of three types of elements: stock elements (also known as state variables), flow elements, and auxiliary variables, along with constants that do not have directional orientation. Auxiliary variables, defined by mathematical formulas, represent the remaining elements, while stock variables are depicted in rectangular boxes. Relationships between various components are shown with arrows, and these relationships can be modified using a window mask. Inflows and outflows of stock variables are represented by double-line arrows [13]. Once the SD model is developed, the simulation can be implemented. Figure 3 provides a detailed overview of these steps.

The fundamental model for this study was established by examining the relationships between key variables, with an emphasis on the primary objectives of the mandatory battery electric vehicle (BEV) policy. This policy seeks to manage the overpopulation of internal combustion engine (ICE) vehicles, which are major contributors to traffic congestion and air pollution while promoting the adoption of BEVs or public transportation for daily commutes. The transformation process is shaped by various factors, including purchasing behaviors, economic growth, population trends, and the availability of supporting infrastructure.

The simulation spans from 2023 to 2060, in line with Indonesia's target year for reaching Net Zero Emissions as specified in Law No. 16 of 2016, which ratifies the Paris Agreement (Republic of Indonesia, 2016). This extended timeframe allows the simulation to capture the long-term effects of the policy and to assess Jakarta's potential for meeting its emission reduction targets comprehensively.

By encompassing the entire period from 2023 to 2060, the study identifies the optimal years for achieving significant emission reductions, as indicated by decreasing air pollution levels. This robust timeframe provides a solid foundation for understanding both the immediate and long-term impacts of the policy, delivering valuable insights for decision-makers and stakeholders involved in Jakarta’s shift towards sustainable urban mobility.

3.2 Data collection

This study relies on diverse, reputable secondary data to ensure reliable inputs for the simulation, drawing on key sources that collectively enhance its robustness and policy relevance.

The Central Bureau of Statistics (BPS) of Indonesia provided vital data, including car ownership by province, population statistics, and regional GDP figures. This data, collected through standardized national surveys and censuses, ensures credibility and consistency when analyzing socioeconomic factors that influence vehicle usage.

Traffic data from the Jakarta Department of Transportation includes road-level service indicators and congestion patterns, delivering essential insights into urban mobility dynamics. Additionally, the Association of Indonesia Automotive Industries (GAIKINDO) offered detailed automotive market data, such as market shares by vehicle type, pricing, and characteristics of internal combustion engine (ICE) and battery electric vehicles (BEVs).

Air quality data were obtained from monitoring reports in Jakarta, supplemented by insights from TomTom Traffic. These datasets provided information on pollutant concentrations (e.g., PM_2.5 and NO_x), trip lengths, and emissions data, facilitating a comprehensive assessment of the relationship between traffic patterns and air pollution.

Energy-related data, including electricity charging rates for BEVs and fuel prices for ICE vehicles, were sourced from PT PLN and PT Pertamina, supplying critical parameters for evaluating the economic feasibility of transitioning to BEVs.

By integrating these diverse datasets, validated through multiple institutions, the study ensures a robust and context-specific simulation that supports actionable policy recommendations for reducing reliance on ICE vehicles, promoting BEV adoption, and enhancing urban air quality.

The primary factors selected for analysis include road capacity, air pollutant concentration, vehicle numbers, population growth, car sales, the number of charging stations, and fuel prices. The total cost of ownership is a significant consideration for individuals in Indonesia when choosing between ICE and BEV vehicles [14]. Each of these factors will be treated as variables in the System Dynamics modeling process.

After identifying the key factors, it is essential to quantify and formalize the relationships among them. Certain mathematical relationships are evident; for example, road capacity, air pollutant concentration, vehicle quantity, population, car sales, number of charging stations, and fuel prices are primarily determined by their initial quantities and growth rates, allowing for direct formulation.

3.3 Problem identification and measurement

3.3.1 Road clustering for policy implementation

Clustering analysis is a method used to examine trip characteristics [15]. In this study, data from 25 road sections were grouped into three clusters: Cluster 1, Cluster 2, and Cluster 3. The primary purpose of clustering the road sections is to categorize them based on similar levels of service or traffic congestion.

Once the road sections are grouped, distinct scenarios can be developed: Scenario 1 for Cluster 1, Scenario 2 for Cluster 2, and Scenario 3 for Cluster 3. Each cluster will illustrate the effects of the mandatory BEV policy and its correlation with the adoption levels of BEVs and public transport, specifically electric buses. These scenarios can also be utilized to assess the impact of the mandatory BEV policy during specific years. The notation for each variable is detailed in Table 1.

Table 1. Cluster scenario

The decline in the population of ICE vehicles will contribute to an increase in BEV adoption; however, some travelers will still opt for public transportation. The timing of policy implementation will influence the pace of ICE reduction. The number of public transport options is expected to rise significantly, particularly due to limitations in adopting BEVs.

In Europe, Low Emission Zones (LEZ) are commonly employed to address issues such as air pollution and traffic congestion. LEZs have been effective in decreasing the number of polluting vehicles while also encouraging greater use of public transportation. Public transportation plays a crucial role in creating sustainable, efficient, and livable cities. Additionally, it serves as a solution to the increased mobility demands driven by population growth in urban areas [16].

3.3.2 Identifying the scale of BEV and ICE

Even though the BEV Mandatory Policy is implemented in only 25 road sections, its effects are expected to influence the entirety of Jakarta. Additionally, the number of vehicles is impacted by factors such as population growth, car prices, the expansion of charging infrastructure, and the prices of electricity and fuel. These variables present common challenges that affect BEV adoption in various countries.

For example, in Poland, the primary issue is the higher cost of BEVs compared to ICE vehicles, which hinders their adoption [17]. Similarly, Luxembourg and Berlin face challenges related to consumer behavior, vehicle market structure, and fuel prices, all of which slow the acceleration of BEVs [18, 19]. In China, increasing BEV adoption can be facilitated through mandatory improvements in fuel quality [20]. Increasing adoption of BEV can be seen from the ratio between BEV and ICE [21]. The growth of both BEVs and ICE vehicles is influenced by numerous factors, as outlined in Table 2.

Table 2. Variables related to adoption of BEV and ICE

3.3.3 Identification of concentration of air pollution

The adoption of ICE vehicles for transportation has a negative impact on air quality. Even with the establishment of new road networks aimed at reducing traffic congestion, there has been little effect on lowering air pollution levels [22]. Additionally, the odd-even restriction policy has not significantly improved air quality in DKI Jakarta [23]. Air pollutants are assessed based on six parameters: PM₁₀, PM_2.5, SO₂, CO, O₃, and NO₂, as detailed in Table 3 [24].

Table 3. Concentration of air pollution year 2023

The differences in air pollutant concentrations across various regions in Jakarta are closely linked to traffic conditions, particularly the level of service (LOS) on the road network. According to the 2023 Final Report of the Traffic Characteristics Survey in Jakarta, areas experiencing poorer traffic performance, characterized by lower average speeds, tend to have higher levels of air pollutants.

For example, West Jakarta, which has the highest pollutant concentration, operates at an LOS of E, with an average speed of just 23.7 km/h. This level of service indicates severe traffic congestion, marked by frequent stop-and-go conditions that lead to inefficient fuel combustion and increased emissions from internal combustion engine (ICE) vehicles.

In contrast, regions like South Jakarta and Central Jakarta exhibit slightly improved traffic conditions, with an LOS of D and average speeds ranging from 28.5 km/h to 31.2 km/h. Although these areas still face congestion, the relatively smoother traffic flow results in lower pollutant concentrations compared to more congested areas.

PM₁₀ and PM_2.5 are derived from emissions from internal combustion engine exhaust gases. The highest concentrations of PM_2.5 and PM₁₀ among transportation modes occur in urban transit, largely influenced by conditions such as vehicles with open windows and doors [25]. The anticipated increase in BEVs is expected to help reduce the effects of global warming [26]. While adopting public transportation is effective in reducing air pollution, the use of private vehicles (ICE) will have a negative impact, as it contributes to higher pollution levels [27]. Implementing stringent pollution controls along with new technologies is an effective strategy for limiting pollution [28].

3.3.4 Identification of growth of BEV to the traffic congestion

The number of vehicles consists of two types: ICE vehicles and BEV. A mandatory policy for BEVs will be implemented to reduce the use of ICE vehicles while simultaneously increasing the adoption of BEVs and other forms of transportation, such as mass transit. The growth of private vehicles, including both BEVs and ICEs, is expected to remain within road capacity limits to prevent traffic congestion. Traffic congestion can have significant economic impacts, including increased travel time, higher fuel consumption, and uncertainty for drivers and passengers [29]. According to the Ministry of Public Works and Housing, the level of service in Jakarta is determined by road capacity, and the maximum capacity can be calculated using Eqs. (1)-(3) [30].

Vactual $=$ Vfree $\times\left(1-\frac{B E V}{\text { Road Capacity }}\right)$            (1)

Vactual $=$ Vtarget            (2)

$B E V_{\max }=\left(1-\frac{\text { Vtarget }}{\text { Vfree }}\right) \times$ Road Capacity                   (3)

Vtargeted=targeted average speed (km/hour)

Vfree=60 km/hour

BEV=amount of BEV (Unit)

BEVmax=Maximum amount of BEV in targeted average speed (Unit)

The relationship between the level of services, average speed, and the maximum capacity can be shown in Table 4.

Table 4. Level of services

The adoption of BEVs will impact road capacity. As the number of BEVs approaches the maximum capacity of roads, it may lead to a reduction in average speed, resulting in traffic congestion. A mandatory BEV policy is expected to provide a more effective solution than a car-free zone policy, which could complicate freight and logistics operations, impacting product competitiveness [31]. Simultaneously, modernizing the aging truck fleet and implementing stricter freight standards, such as those for BEV trucks, can help address air pollution issues [32]. However, it is essential to monitor the growth of BEVs to assess traffic levels in specific road segments.

3.4 Stock and flow diagram

Implementing a mandatory BEVs policy across all clusters, as indicated by the variables ICE C1, ICE C2, and ICE C3, will significantly accelerate the phase-out of internal combustion engines (ICE). Once the policy takes effect in a designated year, it will restrict ICE vehicles from accessing certain road segments, leaving only two options: using public transport or purchasing a BEV. In this context, electric buses will contribute to reducing air pollution and alleviating traffic congestion [33].

The choice between BEVs and ICE vehicles is influenced by several key factors, including price differences and operational costs. Previous research indicates that vehicle price and usage are the most significant determinants, accounting for 32.3% and 28.1% of decision-making factors, respectively [34]. Currently, BEV prices tend to be higher than those of ICE vehicles, suggesting that incentives are necessary to narrow the price gap. Once prices are comparable, the likelihood of customers choosing BEVs is expected to rise. Additionally, tax subsidies play a vital role in encouraging BEV adoption [35].

Operational costs also impact the adoption rates; when BEVs demonstrate lower operating costs than ICE vehicles, it further incentivizes consumers to make the switch. In terms of energy efficiency, BEVs are the only viable alternative to ICE vehicles, especially when compared to hydrogen technology, which experiences energy losses of 47% to 63% during production [36].

The availability of charging stations and the cost of charging can significantly drive the adoption of BEVs. A well-developed charging infrastructure is essential for facilitating the rapid transition from ICE to BEV [37]. Conversely, ICE vehicles have several factors that can help maintain their market presence, such as lower fuel prices and their generally lower purchase price compared to BEVs.

In Jakarta, the Regional Gross Domestic Product (RGDP) and the population size also influence BEV adoption. An increase in RGDP and population typically leads to a greater demand for new vehicles. Additionally, income levels and educational backgrounds significantly impact the likelihood of purchasing a BEV; households that own BEVs earn, on average, 75% more than those with ICE vehicles, and BEV owners are 80% more likely to have completed college than ICE owners [38].

The demand for new vehicles is also linked to air pollution levels. As adoption rates for BEVs and public transportation rise, the average concentration of air pollution is expected to decrease. The growing adoption of battery electric vehicles (BEVs) is anticipated to affect road capacity, highlighting the need to enhance public transportation systems in order to decrease dependence on private vehicles [39]. Evidence from cities around the world shows that integrated and improved public transport systems can effectively alleviate traffic congestion and promote a shift away from private vehicle use.

In Bogotá, Colombia, the TransMilenio Bus Rapid Transit (BRT) and Sistema Integrado de Transporte Público (SITP) have enhanced accessibility by integrating routes, fares, and infrastructure, especially benefiting peripheral areas [40]. Similarly, Medellin's integrated system of trains, trams, BRTs, gondola lifts, and bike-sharing has contributed to reduced traffic and CO₂ emissions [41]. Cities like Bonn, Germany, and Copenhagen, Denmark, further illustrate how expanding public transportation can lessen reliance on private vehicles and mitigate congestion [42, 43].

Technology also plays a crucial role in the adoption of public transport. Factors such as accessibility, availability, and user satisfaction are key influences on travelers' decisions [44]. For example, in China, satisfaction with bus services has been shown to decrease the inclination to own private cars [45].

In addition to technology, integrated regional development that leverages public transportation, like Intercity Rail Transportation exemplified by Greater Jakarta, could also be a key driver in supporting urban sustainability in the future [46].

Figure 4 presents stock-flow diagram of BEV modeling and their correlation with variables.