Pa ge 1 Pa ge 87 American Journal of Smart Technology and Solutions (AJSTS) Smart Transportation Systems with Artificial Intelligence: Enhancing Efficiency, Safety, and Sustainability Abdullah Sheikh1*, Md. Shakil Sheikh2, Tajbiha Mehonaj Rinvee3 Volume 4 Issue 2, Year 2025 ISSN: 2837-0295 (Online) DOI: https://doi.org/10.54536/ajsts.v4i2.6022 https://journals.e-palli.com/home/index.php/ajsts Article Information ABSTRACT Received: August 27, 2025 Accepted: October 03, 2025 Published: November 08, 2025 Artificial intelligence (AI) is transforming transportation, yet most research and applications focus on isolated improvements, lacking a unified approach that connects operational gains with strategic national goals. This paper addresses this gap by developing a conceptual framework that synthesizes how AI enhances transportation systems across three integrated pillars: efficiency, safety, and sustainability. Through a synthesis of recent literature and industry case studies, we propose a model that demonstrates the synergistic effects of AI applications, such as predictive maintenance and dynamic routing. The framework’s primary contribution is to illustrate how these technological advancements collectively bolster U.S. competitiveness by building resilient supply chains, reducing emissions, and fostering leadership in sustainable innovation. This study provides a structured roadmap for policymakers and industry leaders to leverage AI not merely for operational efficiency, but as a strategic asset for long-term economic security. Keywords Artificial Intelligence, Efficiency, Logistics, Safety, Smart Transportation, Sustainability, U.S. Competitiveness 1 Wright State University, Dayton, Ohio, USA 2 Atish Dipankar University of Science & Technology, Bangladesh 3 BRAC University, Bangladesh *Corresponding author’s e-mail: adustabdullah@gmail.com INTRODUCTION Transportation is the backbone of modern economies and defines the movement of goods, services and people in regions and markets. In the United States, the transportation sector supports millions of jobs and contributes greatly to economic growth. At the same time, it is one of the largest sources of carbon emissions and faces growing challenges of congestion, safety and rising operational costs. These pressures underscore the urgent need for intelligent and sustainable transport systems. Artificial intelligence (AI) is a central tool for solving these challenges. Unlike traditional optimization techniques, AI processes large amounts of real-time data, detects complex patterns, and generates adaptive solutions. Applications range from predicting traffic flows and adapting flight routes, improving fleet fuel efficiency, and supporting autonomous vehicle systems. With this combination of efficiency and adaptability, AI is a powerful enabler for change in the transport sector. The relevance of smart transportation goes beyond the operation. It has a strategic impact on national competitiveness and resilience. Countries successfully integrating artificial intelligence into their transportation systems are better equipped to reduce costs, reduce emissions and maintain reliability in the face of disasters such as pandemics, natural disasters or geopolitical tensions. For the United States, which is heavily dependent on timely transportation of goods through its logistics network, intelligent transportation with AI is not only a technological innovation, but also a matter of long-term economic security. This paper focuses on how AI can be applied to improve three basic outcomes in transport: efficiency, safety and sustainability. Efficiency means reducing delays, fuel consumption, and waste through predictive analytics and real-time optimization. Safety is accompanied by computer vision systems, driver monitoring and predictive maintenance with AI. Sustainable development is achieved through reduction in carbon emissions, optimization of energy use and support for the transition to greener transport modes. These three dimensions together form a framework for intelligent transportation driven by AI, supporting the competitiveness and leadership of the United States in sustainable innovation. The remainder of the paper is as follows. In the second section, the literature on AI applications in intelligent transport is reviewed, and both progress and current gaps are highlighted. Section 3 introduces the conceptual framework and methodology. Section 4 presents case studies by industrial leaders. Chapter 5 deals with the impact on the competitiveness of the United States. Finally, section 6 concludes with a summary of the findings and recommendations of future research. LITERATURE REVIEW Evolution of Smart Transportation & ITS Smart transportation systems (STSs) and intelligent transportation systems (ITS) have evolved over the past two decades. ITS integrates perception, communication, calculation and control to improve transport network mobility, safety and environmental performance. Zemmouchi-Ghomari et al. (2025) Review how AI is integrated into ITS to support traffic flows, safety and sustainability in urban areas. Recent research has also highlighted the influence of generational AI within ITS. For example, Rong and others. (2025) Examine applications such as data generation, prediction and decision-making in ITS subsystems. Pa ge 88 https://journals.e-palli.com/home/index.php/ajsts Am. J. Smart. Technol. Solutions 4(2) 87-90, 2025 In recent research, sustainability and efficiency have become increasingly closely linked. Son et al. (2025) conducts a systematic review to demonstrate how AI, IoT, digital twins and optimization methods are used in intelligent transportation planning, highlighting important improvements in traffic flows and emission reductions. AI Methods Applied in Transportation AI methods used in transportation are diverse. They include: • Machine Learning (ML) and Deep Learning: For traffic prediction, demand forecasting, incident detection, and routing decisions. • Reinforcement Learning (RL): It is used to adapt control strategies dynamically (e.g. traffic signal control, dynamic routing). Li et al.’s bibliographic review. (2022) shows that RL is increasingly attractive in transportation applications. • Generative Models/Generative AI: Emerging for tasks such as creating synthetic data, simulation of scenarios, or improving prediction under data-sparity conditions (for example, Yan and Li, 2023). • Hybrid Approaches: Combining AI with IoT sensors, blockchain and optimization. Idrissi et al. (2024) Examine how IoT, IoT, and Blockchain interplay in logistics and transport, and improve the availability, resilience, and support for decision-making. Key Application Areas & Findings AI in smart transportation is applied to many problem domains. Here are several with findings: • Traffic prediction & routing: Many studies show that AI is more powerful than traditional models (such as ARIMA and linear regression), especially in nonlinear and volatile traffic environments. • Safety & accident prediction: Computer vision and sensor-based ML detect risky driving behaviors or predict likely crash spots. • Predictive maintenance: Using sensor data and ML to anticipate vehicle or infrastructure failures before they happen. • Energy optimization & emissions: AI optimizes route planning, load balancing, and speed profiles to reduce fuel use and emissions. • Mobility-as-a-Service (MaaS): Rouky et al. (2025) Examine how AI supports integrated mobility systems (routes, payments, user behaviors) and link AI technologies to integrated levels. Challenges, Gaps, and Research Needs Despite progress, several gaps and challenges persist: • Data quality and heterogeneity: Many models struggle when input data are missing, noisy, or inconsistent. • Real-time requirements: AI models must be fast and efficient to operate in real-time constraints. • Scalability: Many prior works focus on small regions or simulated settings; scaling to national networks is harder. • Integration across systems: Too often, applications are isolated (only traffic or routing). Few studies integrate safety + emissions + efficiency within a single framework. • Ethical, privacy, and security concerns: Use of AI in transportation brings risk of surveillance, data breaches, fairness issues. • Lack of longitudinal studies: Few works track AI performance over long time or in real operations (versus simulations). How This Paper Contributes Our work addresses these gaps. We propose a unified framework for efficiency, safety, and sustainability in intelligent transportation. Unlike many previous studies, we emphasize the coherence of the real world over the application domain rather than isolated models. It is also intended to highlight the strategic implications for US competitiveness, which are under-emphasized in the current literature. MATERIALS AND METHODS This paper uses conceptual and exploratory methods rather than empirical methods. The objective is to develop a framework to link artificial intelligence (AI) to intelligent transportation, efficiency, safety and sustainability, while focusing on national competitiveness. Research Approach The methodology is based on comparative synthesis of recent studies, reports and industry practices. Instead of analyzing a single dataset, we reviewed academic and industrial literature published in the fields of transportation, logistics and AI between 2018 and 2025. The aim was to identify repeated applications, evaluate their results and organize them into coherent models of smart transportation. Data Sources The source of this study is: • Articles reviewed by peer-reviewed professionals on AI in transportation, logistics and sustainability. Industry reports from major transport companies such as Amazon, UPS, Uber Freight and Tesla. • Policy documents from American and international organizations relating to intelligent cities, clean transportation and digital infrastructure. Using various sources ensures that the framework is based on both academic theory and practical evidence. Framework Development The framework was developed in three steps: 1. Mapping Applications: Key applications of AI in transportation, such as traffic forecasting, routing, predictive maintenance, safety, energy optimization, and emissions monitoring, were identified and categorized. 2. Identifying Outcomes: For each application, the dual outcomes were noted: efficiency gains (reduced costs and faster operations) and sustainability benefits (reduced Pa ge 89 https://journals.e-palli.com/home/index.php/ajsts Am. J. Smart. Technol. Solutions 4(2) 87-90, 2025 emissions and improved safety). 3. Linking to Competitiveness: Finally, we linked these outcomes to national competitiveness by arguing that countries with stronger AI-driven transportation systems gain resilience and leadership in the global economy. Scope and Limitations This is a conceptual study, not an empirical one. While it provides a broad synthesis, it does not include primary data collection or quantitative testing. However, the paper offers a roadmap for both researchers and policymakers by organizing existing knowledge into a unified model. The Proposed Framework for AI in Smart Transportation This section presents the framework for demonstrating how artificial intelligence supports the efficiency, safety and sustainability of smart transportation systems. The framework is based on three pillars: operational efficiency, road safety and environmental sustainability. These pillars together strengthen the long-term competitiveness of the country. Figure 1: Conceptual Framework “AI in Smart Transportation” This figure illustrates how AI applications strengthen efficiency, safety, and sustainability, which together reinforce U.S. competitiveness. Efficiency through AI AI reduces inefficiency across transport networks. Analyzing real-time traffic data, AI systems adjust signal timing, redirect vehicles, and predict traffic congestion before it occurs. In the logistics sector, platforms such as Uber Freight and Amazon’s routing system allocate trucks more effectively, reducing empty miles and reducing costs. Predictive maintenance also supports efficiency. Machine learning models monitor vehicle health and forecast component failures, allowing rapid repairs and reducing shutdown times. Safety Improvements Safety is the second pillar of the framework. AI-enabled systems, such as driver assist technology, Lanekeeper system, and automatic brake, reduce the risk of accidents. AI can also process real-time road, weather and driver data and emit warnings before the situation escalates. For example, predictive analytics can identify dangerous driving patterns and trigger alerts or training interventions. These applications not only save lives but also reduce the costs of accidents financially and socially. Sustainability and Emission Reduction Artificial intelligence directly contributes to sustainable development by reducing energy consumption and emissions. Dynamic routing reduces fuel consumption by selecting the most efficient path. Electric vehicle fleets can be managed by AI to forecast charging requirements and align them with renewable energy supply. Companies like Walmart and UPS already use artificial intelligence to reduce energy consumption in warehouses and optimize fleet management to reduce emissions. At the city level, intelligent traffic lights based on AI reduce idle time and cause measurable carbon production reductions. Linking to Competitiveness Combining the three pillars of efficiency, safety, and sustainability creates a competitive advantage. The nation and company adopting AI in transport gains resilience to disruption, better compliance with environmental standards and more reliable supply chains. For the United States, this not only means reducing costs, but also enables sustainable innovation, develops exportable technologies and strengthens its position in the global market. RESULTS AND DISCUSSION The proposed framework emphasizes that artificial intelligence is not a set of digital tools. It is a strategic driver of efficient, safe and sustainable transport. By combining these pillars, artificial intelligence offers a path to strengthening national competitiveness. From the business point of view, the framework shows that artificial intelligence investments can create double value. Companies benefit financially by reducing costs and delays, while improving social results such as accident reduction and emission control. This dual impact makes it easier to justify AI adoption to stakeholders. From a policy point of view, the framework provides a roadmap for governments to develop regulations and incentives to support them. For example, subsidies to electric vehicles equipped with artificial intelligence can reduce emissions, while data sharing policies can improve traffic management throughout cities. Policy makers can also encourage public and private partnerships to accelerate the deployment of AI solutions. From a research point of view, the framework identifies gaps in future research. It is necessary to measure the long-term impact of the adoption of artificial intelligence, particularly how efficiency is balanced with sustainable outcomes. More empirical research is also needed on the social effects of AI-based safety systems, especially in areas with high incidence of accidents. Pa ge 90 https://journals.e-palli.com/home/index.php/ajsts Am. J. Smart. Technol. Solutions 4(2) 87-90, 2025 Overall, the implications of this framework show that leading companies and countries in the field of AI- enabled transportation will not only reduce costs and risks, but also become leaders in sustainable innovation. For the United States, this offers the opportunity to strengthen resilience, maintain technological leadership and enhance global competitiveness. Over the next decade, the most important research needs will be the testing of AI models not only for their efficiency, but also for their impact on fairness and long- term sustainability, particularly in the transportation networks of the United States. CONCLUSION This paper developed a conceptual framework establishing AI as a catalyst for integrated gains in transportation efficiency, safety, and sustainability. The key insight is that the synergistic effect of these pillars is crucial for building a resilient and competitive national transportation system. For U.S. competitiveness, this means policymakers should prioritize funding for integrated AI projects that cross environmental and operational agencies. Industry leaders must move beyond point solutions and adopt platforms that unify these three objectives. For researchers, the critical path is to generate empirical evidence on the framework’s synergies in real-world settings and to establish standards for ethical AI deployment. Ultimately, strategic adoption of this unified AI framework is not optional but essential for the U.S. to secure economic strength and global leadership in the future of mobility. REFERENCES Chen, C., Demir, E., Huang, Y., & Scholts, S. (2022). AI and big data in sustainable transportation: Opportunities and challenges. Transportation Research Part E: Logistics and Transportation Review, 159, 102620. https://doi.org/10.1016/j.tre.2021.102620 Choi, T. M., Chen, Y., & Lee, W. W. (2018). Big data analytics in supply chain management: A review and research agenda. Journal of Management Information Systems, 35(2), 528–567. https://doi.org/10.1080/07 421222.2018.1440770 Dwivedi, Y. K., Hughes, L., Kar, A. K., Baabdullah, A. M., Grover, P., Abbas, R., & Mani, V. (2021). Climate change and COP26: Are digital technologies and information management part of the problem or the solution? International Journal of Information Management, 63, 102456. https://doi.org/10.1016/j. ijinfomgt.2021.102456 Ferreira, A. C. A., Francisco, M., & Pinho, A. (2025). The use of artificial intelligence in transportation and logistics: A systematic literature review. IEEE Access, 13, 1–14. https://doi.org/10.1109/ ACCESS.2025.3275890 Hasan, M. R., Islam, M. R., & Rahman, M. A. (2025). Developing AI-driven models for demand forecasting in U.S. supply chains: Enhancing predictive accuracy. Edelweiss Applied Science and Technology, 9(1), 1045–1068. Ivanov, D. (2020). Predicting the impacts of COVID-19 disruptions on global supply chains: A simulation- based analysis. International Journal of Production Research, 58(20), 6140–6156. https://doi.org/10.1080 /00207543.2020.1750727 Lee, J. (2024). AI-powered forecasting in transportation: Accuracy, speed, and scalability. Multidisciplinary Journal of Instruction, 7(1), 115–125. McKinsey & Company. (2023). The future of mobility: AI- driven sustainability in transportation. McKinsey Insights. https://www.mckinsey.com Pattnaik, S., Liew, N., Kures, A. O., Pattnaik, E., & Park, K. (2024). Catalyzing smart transportation: AI applications for sustainable logistics. Engineering Proceedings, 68(1), 57. https://doi.org/10.3390/ecsa- 68-057 Waller, M. A., & Fawcett, S. E. (2013). Data-driven supply chains: A new lens on supply chain management. Business Horizons, 56(5), 639–647. https://doi. org/10.1016/j.bushor.2013.06.001 Wieland, A., & Marcus, F. (2020). The role of dynamic capabilities in responding to supply chain disruptions. International Journal of Production Research, 58(10), 2904– 2915. https://doi.org/10.1080/00207543.2020.1744 519