Artificial Intelligence-Based Intelligent Navigation System for Alleviating Traffic Congestion: A Case Study in Batam City, Indonesia

Rapid population growth and uncontrollable urbanization have become a global phenomenon that significantly impacts transportation management in urban areas. Many cities around the world, both metropolitan and medium-sized cities that are rapidly developing, are facing increasingly complex traffic congestion challenges [1-3]. This congestion not only causes discomfort for road users but also negatively affects various aspects of life, such as increased travel time, excessive fuel consumption, rising carbon emissions, and reduced economic productivity due to delays in the mobility of people and goods [4, 5]. In the global context, the World Health Organization (WHO) and the United Nations Human Settlements Programme (UN-Habitat) have highlighted that dense urban traffic contributes significantly to air pollution and a decline in the quality of life for citizens [6, 7]. Major cities around the world, such as Jakarta, Manila, Bangkok, and Mumbai, serve as real-life examples of how population pressure and the growth of motor vehicles can exceed the capacity of existing infrastructure, resulting in mobility stagnation. To address this, many countries have developed Intelligent Transportation Systems (ITS), which are technology and information-based transportation systems aimed at improving traffic efficiency and safety [8].

In Indonesia, traffic congestion issues are not only prevalent in large cities like Jakarta, Surabaya, and Bandung but are also increasingly being felt in medium-sized cities that are growing economically, one of which is Batam. Batam is a strategic area located directly at the border with Singapore and Malaysia and is part of a special economic zone promoted by the Indonesian government [9]. The growth of industry, trade, and tourism in Batam has led to a surge in population and vehicle numbers over the past few decades. According to data from the Central Statistics Agency (BPS) and the Batam City Transportation Department, the growth of motor vehicles in Batam has increased by more than 8% per year, mostly dominated by private vehicles such as motorcycles and cars [10, 11]. The increase in vehicle volume is not accompanied by an adequate increase in road capacity [12, 13]. Major roads in Batam, such as Jalan Sudirman, Jalan Yos Sudarso, and Jalan Ahmad Yani, often experience traffic congestion, especially during peak hours in the morning and evening [14, 15]. This phenomenon indicates an imbalance between the demand and the available transportation service capacity. Additionally, the uneven distribution of residential and industrial areas further complicates traffic movement patterns in the city [16]. This congestion not only hampers public productivity but also increases the risk of traffic accidents and air pollution [17].

The Batam City Government’s efforts in managing traffic still largely rely on traditional approaches, such as static traffic light timing, the construction of alternative roads, and the implementation of road signs and markings to control road user behavior [18]. While these approaches are important and fundamental, they have proven to be insufficiently responsive in addressing the highly fluctuating traffic dynamics in the modern era. Limitations in real-time traffic monitoring systems and the suboptimal integration of traffic data from various sources make decision-making slow and less adaptive [19].

GPS-based navigation systems commonly used by drivers mostly rely on conventional route-finding algorithms that do not consider real-time traffic conditions [20]. As a result, road users are often directed to shorter routes that are heavily congested, thereby reducing the effectiveness of navigation [21]. This highlights the need for a transformation towards a more intelligent, adaptive, and data-driven system to support more efficient traffic management [22, 23].

Artificial Intelligence (AI) has become one of the key technologies in the development of Smart Cities, including in the field of transportation [24, 25]. With its ability to analyze large amounts of data, learn patterns, and make decisions automatically and adaptively, AI offers great potential to be applied in modern transportation systems. In the context of traffic, machine learning algorithms and reinforcement learning have been used for traffic flow prediction, signal timing optimization, and the selection of the fastest routes based on real-time data [26, 27]. The use of Design Science Research Methodology (DSRM) in this domain allows for a structured approach in the design, development, and evaluation of AI-based solutions. Through iterative cycles of artifact design and evaluation, DSRM ensures that the AI systems developed are not only effective but also relevant to the specific challenges of urban traffic management. By leveraging DSRM, this research aims to create an adaptive, data-driven navigation system that can optimize traffic management in Smart Cities like Batam.

AI-based navigation systems offer advantages over conventional approaches as they can learn from previous traffic patterns and dynamically respond to changes in traffic conditions. For example, if an accident or obstruction occurs on a certain road, the system can immediately redirect drivers to alternative routes that are less congested. In various developed countries, these systems have been integrated with traffic sensors, CCTV cameras, and user application data to create a connected and adaptive transportation ecosystem [28, 29]. Although the implementation of AI-based intelligent navigation systems has developed rapidly in large cities with advanced technological infrastructure, there remains a significant gap in the application of this technology in developing cities like Batam. Most existing studies and developments are designed for large cities with extensive data access, complex sensor networks, and high computational resources. In fact, medium-sized cities also face the same urgent needs to optimize traffic, albeit with limited infrastructure and data [30].

Batam, with its unique characteristics, requires a tailored approach and cannot simply adopt systems used in advanced cities [31-33]. This approach must consider differing road conditions, uneven vehicle distribution, and the limitations of integration between transportation information systems. Local innovation is crucial in designing navigation systems that can work effectively with limited resources while still leveraging the power of AI technology [34, 35].

This study successfully developed an AI-based intelligent navigation system to address traffic congestion in Batam City. The system proved effective in reducing average travel time by 22.8% compared to conventional navigation systems, as well as distributing traffic load more evenly across the road network. Additionally, the system provides adaptive rerouting based on real-time traffic conditions, allowing users to select more efficient routes.

The route prediction accuracy of the AI system reached 91.3%, higher than the 85.6% accuracy of conventional GPS systems, demonstrating the superiority of the AI system in optimizing travel. The rerouting time of this system is also very fast, with an average latency of only 423 milliseconds, enabling users to receive route directions promptly after changes in traffic conditions occur.

User experience with the system has been highly positive, with a high satisfaction rate, indicating strong public acceptance of this technology. This reflects that the system is not only effective in improving traffic efficiency but also provides a more comfortable and safer driving experience for users.

Overall, this AI-based navigation system offers an effective solution for reducing congestion, improving transportation efficiency, and optimizing traffic management in Batam. This technology can be applied to other developing cities with adjustments to local conditions, supporting sustainable development goals such as reducing carbon emissions and fuel consumption. Moving forward, integrating this system into city transportation policies could have a broader positive impact on the quality of life and traffic safety.