The paper introduces an intelligent logistics system which uses machine learning together with heuristic routing algorithms to enhance last-mile delivery efficiency. The system uses a Random Forest regression model to predict customized delivery time slots which it trains with distance traffic weather and delivery priority data. The A* (A-Star) search algorithm enables route optimization through its ability to determine the shortest path by using heuristic distance calculations. The system aims to reduce delivery delays while decreasing operational expenses and increasing customer contentment through its ability to match delivery times with current operational conditions. The system uses FastAPI to create backend services while React handles the frontend interface and PostgreSQL stores data in its modular full-stack architecture. The experiments show that our system achieves better efficiency compared to existing systems which use fixed routing and scheduling methods. The proposed approach highlights the potential of combining predictive analytics with heuristic optimization in modern logistics applications. TheYou system uses FastAPI to create backend services while React handles the frontend interface and PostgreSQL stores data in its modular full-stack architecture.

The AI-based customized time slot delivery system together with the route optimization system provides complete solutions for the present-day difficulties encountered by logistics and delivery operations. The system uses machine learning together with heuristic routing methods to solve scheduling problems and delivery estimation issues and routing efficiency challenges. The Random Forest regression model analyzes various delivery time factors to provide accurate delivery time predictions while the A* algorithm uses heuristic search methods to calculate routes. The system achieves scalable performance through its modular design which utilizes API-based communication to connect with external services. The system achieves better usability through its web-based dashboard which allows users to interact with the system in real time while providing them with an enhanced user experience. The proposed solution shows better operational results than standard delivery systems because it delivers higher efficiency ratings and better adaptability and improved overall performance. The system has benefits but it needs to overcome three specific difficulties which include its need for data, its limitations in scaling, and its requirement for receiving current data. The existing problems can be solved through upcoming scientific studies and developments in technology. The proposed system brings major progress to intelligent logistics systems while it will revolutionize the delivery sector through its improved operational efficiency and reliability and customer-focused approach.

[1]Wang Zhang Liu present their research on machine learning delivery time prediction for smart logistics systems in their paper published by IEEE Access which appears in volume 10 and extends from page 45231 to page 45245 in the year 2022. [2]Sharma Gupta present their research which compares different regression models used for logistics prediction systems in their article published in Journal of Big Data which appears in volume 9 and issue 2 from page 1 to page 18 in the year 2023. [3]Patel Mehta developed an artificial intelligence routing system which uses heuristic algorithms for route optimization in their work published in IEEE Transactions on Intelligent Transportation Systems which appears in volume 23 and issue 6 from page 5123 to page 5135 in the year 2022. [4]Chen and others developed an A* algorithm which provides efficient performance for real-time navigation systems in their research published in IEEE Access which appears in volume 11 from page 12567 to page 12580 in the year 2023. [5]Kumar Verma demonstrate how artificial intelligence enables dynamic scheduling for smart delivery systems in their research published in Future Generation Computer Systems which appears in volume 130 from page 112 to page 125 in the year 2022. [6]Nguyen Tran present their research on adaptive delivery systems which utilize real-time data integration in their article published in Computers and Industrial Engineering which appears in volume 169 in the year 2023. [7]Zhou and others present a comprehensive review on artificial intelligence applications through their research which explores emerging trends in logistics from their work published in IEEE Access which appears in volume 9 from page 162345 to page 162360 in the year 2021. [8]Singh Kaur demonstrate how deep learning techniques can improve smart logistics systems and supply chain operations in their research which appeared in Expert Systems with Applications volume 198 in 2022. [9]Roy Das present their research on how artificial intelligence technologies enhance decision support systems for transportation systems in their study published in Transportation Research Part C which appears in volume 144 during the year 2023. [10]Lee Park present their research on optimization techniques which use heuristics to enhance routing systems in their article published in IEEE Systems Journal which appears in volume 17 and issue 2 from page 2024 to page 2035 in the year 2023. [11]Fielding and others present their research on designing modern RESTful APIs which enable scalable web system architecture in their study published in ACM Computing Surveys which appears in volume 55 and issue 4 during the year 2022. [12]Dean Ghemawat developed a system which allows large-scale applications to operate across multiple distributed components in their research published in Communications of the ACM which appears in volume 65 and issue 7 from page 68 to page 78 in the year 2022. [13]The OpenAI Technical Report published in 2023 presents research about AI system development and their practical uses. [14]Kaggle provides datasets about logistics and delivery time estimation through their 2023 resource. [15] Google Maps Platform Documentation shows developers how to use Google Maps Platform according to Google documents from 2024. [16]The PostgreSQL Global Development Group published PostgreSQL documentation which they released in 2024. [17]The FastAPI Framework Documentation describes the framework according to Sebastián Ramírez. [18]Meta provides its React framework documentation through the Meta documentation for 2024. [19]The NetworkX Developers created NetworkX Graph Library Documentation which they published in 2024. [20]The Scikit-learn Developers created Machine Learning in Python documentation which they published in 2024. [21]M. Hassan A. A. Nafees S. S. Shraban A. Paul H. D. Mahin together with their article “Application of Machine Learning in Intelligent Transportation Systems: A Comprehensive Review and Bibliometric Analysis” published in Discover Civil Engineering volume 2 issue 98 in the year 2025. X. Liu Y.-L. Chen L. Y. Por C. S. Ku wrote their article A Systematic Literature Review of Vehicle Routing Problems with Time Windows which appeared in Sustainability volume 15 issue 15 of 2023.

© 2025 International Journal of Engineering and Techniques (IJET).

Digital Object Identifier (DOI)

IJET assigns a unique DOI to every accepted article via Zenodo for persistent linking and citation. Your article’s DOI: https://doi.org/{{doi}}

Open DOI About Zenodo DOI

Close