Real-Time Route Optimization in Logistics: A Deep Learning Approach

<div class="section abstract"><div class="htmlview paragraph">As electric trucks become more central to modern logistics, the need for smarter, more adaptive route planning is growing rapidly. This paper presents a key navigation feature for analyzing and recalibrating such optimized routes in real time. Integrating map features into the navigation mode improves user experience by offering real-time navigation and dynamic route adjustments based on traffic updates, road closures, vehicle coordinates and deviation in expected energy consumption. This study compares the performance of Server sent events (SSE), web sockets, and Application programming interface (API) polling methodologies, focusing on metrics such as data transmission efficiency, latency, resource utilization, scalability, and reliability. Our results demonstrate the advantages and limitations of each method, providing insights into their suitability for real-time route optimization in electric truck logistics. The results highlight the potential of SSE in achieving efficient and timely data updates, contributing to more effective route planning and resource management. Additionally, we discuss how API Polling, Web sockets, and SSE each make sense in different scenarios when creating a navigation system (drive mode), considering factors such as the frequency of updates, network conditions, and system architecture. This research underscores the importance of choosing the right communication protocol and integrating advanced map features to enhance the performance and reliability of logistics systems.</div></div>

Decentralized last-mile delivery is a modern logistics approach that optimizes the delivery process by using local hubs or microhubs located near the final recipients. This reduces the distance between the hubs and the delivery location, which reduces transportation time and operating costs. At the same time, customers receive their orders faster, which increases their satisfaction with the service. This approach is extremely relevant amid the growing popularity of e-commerce, where speed and accuracy of delivery are crucial to maintaining customer loyalty. However, the successful operation of a decentralized delivery model requires a sophisticated logistics management system. Artificial intelligence technologies come to the rescue to solve a number of tasks, such as route optimization, inventory management, and demand forecasting. For example, machine learning algorithms are able to adapt courier routes in real time to changes in traffic, weather conditions, and road conditions, which reduces fuel costs and reduces the number of vehicles on the road. This is especially important in large urban areas with heavy traffic, where even small delays can significantly affect the overall delivery time. AI also plays an important role in inventory management, allowing for more accurate planning of replenishment of goods in individual local hubs, avoiding excessive accumulation of products or their shortage. By analyzing historical sales data and consumer behavioral patterns, AI systems can predict the demand for different categories of goods in certain areas. This helps minimize storage costs and ensure that products are available in the right place and at the right time, which in turn increases delivery efficiency. Thus, a decentralized last-mile delivery model involving artificial intelligence allows logistics companies to significantly improve the quality of service, increase the efficiency of all logistics processes, and reduce the negative impact on the environment.

Traffic congestion has become a serious social problem in many countries and cities. In recent years, with the popularization of car navigation systems and the advancement of traffic information provision technologies, there has been an increasing demand for more accurate congestion prediction. In particular, studies have shown that the use of machine learning for congestion prediction is effective, as it can take into account various factors affecting traffic volume, such as weather, events, and days of the week. In addition, existing research has confirmed that it is possible to predict congestion patterns in advance based on past traffic data. However, studies have pointed out that existing machine learning models struggle to ensure sufficient prediction accuracy for abrupt changes in traffic flow caused by sudden events such as traffic accidents. Since the occurrence of accidents is accompanied by discontinuous fluctuations, general time-series prediction models cannot appropriately reflect their impact, leading to a tendency for larger prediction errors. Therefore, there is an urgent need to develop prediction methods that can adapt to sudden changes in traffic conditions. To address this issue, this study aims to construct a machine learning model using multivariate LSTM (Long Short-Term Memory) to achieve more accurate congestion prediction. LSTM excels at capturing long-term dependencies in time-series data and is adept at making predictions based on past data. Thus, it is highly effective in learning the occurrence trends of congestion with certain patterns. Specifically, the purpose of this study is to analyze the trends of prediction errors when traffic accidents occur and to construct a congestion prediction framework that considers the impact of accidents. In this way, we strive to realize a congestion prediction model that can adapt to sudden changes in traffic conditions, contributing to the optimization of traffic flow and the alleviation of congestion.

The metro station can form a new commercial district.The residential development in the outskirts has a major impact on the value of real estate.However, the area near the metro station can be wrongfully affected due to the structure and function of the railway station, noise or congestion, etc.This study aims to analyze how changes in traffic conditions affect the apartments.The opening of the urban road is considered as traffic change.Regarding it, the price trend will be investigated.There is an assumption that preference will grow when opened.Daegu Subway Line 3 in Korea, which opened in 2015, is selected.It is the relatively recent case.Sales price of apartment houses included in the station area and apartment houses outside the station area is obtained, from which we calculate the rate of change.The difference between station area and non-station area was analyzed under two hypotheses whether it existed or not.The 2015 case and after-one-year case were compared.The subway station proved to raise housing price.T-test was utilized to confirm the reliability of the results.They were significant at the 95% confidence level of the two-sided test.Consequently, the station apparently tends to increase public housing price.

Urban transportation systems face increasing challenges due to rapid population growth, traffic congestion, and unpredictable road conditions. Traditional routing algorithms like Dijkstra and A* are limited in their ability to respond to real-time events such as accidents, roadwork, or weather disruptions. This study aims to develop a smarter, more adaptive route optimization system using machine learning techniques. The goal is to enhance travel time accuracy, reduce congestion, and improve commuter satisfaction through intelligent, data-driven decision-making. The proposed method integrates supervised learning for travel time prediction and reinforcement learning for real-time route selection, using data from GPS trajectories, traffic flow, weather reports, and user behaviors. A grid-based environment is used for reinforcement learning simulations, while OpenStreetMap data supports city-level route optimization. Results show that the machine learning-enhanced model significantly outperforms traditional algorithms in terms of adaptability, responsiveness, and reliability. In particular, reinforcement learning proved effective in dynamic environments, learning optimal routes over time and adjusting to disruptions. This research contributes to the development of intelligent transportation systems and supports the broader vision of smart cities, where mobility is safer, faster, and more efficient through the power of AI and real-time data integration.

The last-mile delivery problem is one of the most complex and resource-intensive aspects of modern logistics, especially within the growing e-commerce sector. As online shopping continues to expand, companies are under immense pressure to deliver goods more quickly, efficiently, and at lower costs, all while meeting the demands of increasingly time-sensitive customers. This has created a need for innovative solutions that can tackle challenges related to dynamic traffic patterns, fluctuating customer preferences, and operational constraints such as vehicle capacities and delivery windows. In response to these challenges, this paper explores the application of predictive analytics as a tool for optimizing last-mile delivery routes in real-time.The study begins by identifying the core challenges inherent in last-mile logistics, particularly in the U.S. e-commerce landscape, where the cost of last-mile delivery can represent up to 53% of total shipping costs. With traffic congestion, unpredictable customer availability, and delivery time constraints posing significant hurdles, conventional static route planning models often fall short. In this paper, predictive analytics is proposed as a solution to these challenges, utilizing real-time data to inform more efficient routing decisions. By processing vast amounts of real-time traffic data, customer preferences, and delivery constraints, predictive models can offer a more flexible and responsive approach to last-mile delivery.The research then presents a comprehensive literature review of existing route optimization methods, such as the traditional Vehicle Routing Problem (VRP) and its extensions, including VRP with Time Windows (VRPTW), Dynamic VRP (DVRP), and Capacitated VRP (CVRP). While these models have proven useful, their limitations are exposed when faced with real-time operational complexities in the e-commerce sector. Therefore, this study introduces an advanced dynamic routing model that integrates machine learning algorithms—such as decision trees and neural networks—with traditional VRP frameworks. These machine learning models, trained on historical data, are capable of predicting future traffic patterns, customer behavior, and delivery time windows.A case study is conducted using data from U.S.-based e-commerce companies to demonstrate the practical application of predictive analytics in optimizing last-mile delivery. The case study outlines how predictive models are used to dynamically adjust delivery routes based on real-time conditions, leading to significant improvements in efficiency, cost savings, and customer satisfaction. Key performance indicators such as delivery times, fuel consumption, and vehicle utilization are examined before and after the implementation of the predictive models, with the results showing a reduction in delivery time by 20% and fuel costs by 15%, alongside improved on-time delivery rates.The paper concludes by presenting the proposed dynamic route optimization model as a solution that combines the flexibility and responsiveness of predictive analytics with the robust framework of traditional VRP models. Through the integration of machine learning, real-time data processing, and dynamic routing, the model is shown to significantly improve last-mile delivery efficiency. This study's findings highlight the potential for predictive analytics to revolutionize the logistics industry, particularly in the high-demand e-commerce sector, where quick and reliable delivery is paramount. The research suggests that as e-commerce continues to grow, predictive analytics will play an increasingly critical role in ensuring that last-mile delivery is both cost-effective and responsive to the evolving needs of consumers.

Currently, the transport and logistics business environment are characterized by increasing uncertainties. Last-mile delivery strategies have emerged as important approaches for enterprises to achieve profit and market share objectives by reducing environmental risk experienced. The primary aim of this study was to determine the influence of last-mile delivery strategies on the performance of logistics firms in Mombasa County, Kenya. The research was based on specific objectives, such as route optimization, flexible delivery, green delivery, and crowd-sourced delivery on the performance of logistics firms in Mombasa County, Kenya. Theories, such as the network optimization theory, dynamic capabilities theory, institutional theory, and agency theory, also guided the study. This study employed a cross-sectional survey research design, mainly concentrating on 710 logistic firms in Mombasa County, Kenya. To meet the research objectives, the departmental heads in the transport and logistics department were used as observation units across the firms. The pilot study targeted 25 respondents, 10% of the final study sample size, of which respondents were randomly chosen in each firm. The primary data was collected through self-administered questionnaires that included open and closed-ended questions designed to obtain definite responses. For this study, the quantitative data collected was analyzed by using SPSS version 28 to calculate the response rate using descriptive statistics, including frequencies, percentages, means, and standard deviations. The study findings showed that last-mile delivery strategies namely route optimization, flexible delivery, green delivery and crowd-sourced delivery significantly influence the performance of logistics firms in Mombasa County, Kenya. The study concluded that performance was predicted to improve for every unit increase in these practices. The study further recommended that logistics firms should prioritize having real-time traffic data integrated into their route mapping process. To achieve this, companies should collect up-to-date traffic information from multiple sources, process it through AI-powered route optimization algorithms, continuously monitor changing conditions, and dynamically adjust routes in real time to minimize delays and improve delivery efficiency. This will ensure improved last-mile delivery strategies and improve performance of logistics firms.

Background:The study investigates the emerging concept of Last Mile Logistics (LML), focusing on optimization routing for e-commerce deliveries.This research addresses the growing demand for efficient delivery systems driven by the rise of online shopping and technological advancements.Objective: This article aims to analyse existing literature, identify trends and highlight gaps in LML research.Methods: A systematic approach was used for the selection of studies and then bibliometric analysis was used to map and evaluate relevant studies.The Scopus citation database was used to gather peerreviewed articles from 1999 to 2023.R-Studio and VOS Viewer were employed for data analysis and visualization.The initial search yielded 3,898 articles, which were filtered down to 1,606 articles.Key findings include the identification of the most productive journals, top-cited articles and influential keywords in LML research. Results:The study significantly emphasizes variables such as distance, costs, energy consumption and delivery time in logistics optimization.E-commerce emerged as a predominant topic, with substantial discussions on technology and transportation within LML.The research underscores the critical role of LML in the success of e-commerce. Conclusion:The findings highlight the need for ongoing innovation and optimization in delivery logistics to enhance efficiency and meet consumer demands.Future research should address the gaps identified, particularly in the context of emerging technologies such as autonomous vehicles and drones, to further improve last mile delivery systems.

PurposeThe purpose of this paper is to investigate the effect of a dimension of logistics service quality (delivery time) interacting with shipping charges and purchase importance on customer satisfaction and purchase intentions in an e-commerce context. Uncertainty in terms of perceived ambiguity and perceived riskiness is shown to be the theoretical mechanism that plays a mediating role in the relationships between delivery time and customer satisfaction, as well as between delivery time and purchase intentions.Design/methodology/approachThis study used a scenario-based role playing experiment. Three variables are manipulated in the design of the study – delivery time, shipping charges, and purchase importance. Participant responses (n=360) were collected through Amazon Mechanical Turk with perceptual measures.FindingsResults indicated that increased delivery time significantly increased customers’ perceived ambiguity and perceived riskiness which reduced satisfaction as well as negatively impacted purchase intentions. Further, free shipping reduced customers’ perceived ambiguity when delivery time was lengthy, but strengthened the perception of ambiguity when the delivery time was short.Originality/valueThis paper sheds light on how a dimension of logistics service quality (delivery time) interacts with shipping charges and purchase importance to impact customer satisfaction and purchase intentions. It introduces uncertainty in the form of perceived ambiguity and perceived riskiness, to the logistics service literature as the mechanism that can explain how delivery time interacting with shipping charges and purchase importance impact customer satisfaction and purchase intentions. The implications for online retailers are that they should display separate shipping charges for shorter delivery times but for longer delivery times they should display a total price for the product which includes the shipping cost. Also when the purchase is important to the customer, they should offer shorter shipping time choices if they want to increase customer satisfaction.

The flying sidekick traveling salesman problem with stochastic travel time: A reinforcement learning approach

Machine learning offers advanced tools for efficient management of radio resources in modern wireless networks. In this study, we leverage a multi-agent deep reinforcement learning (DRL) approach, specifically the Parameterized Deep Q-Network (DQN), to address the challenging problem of power allocation and user association in massive multiple-input multiple-output (M-MIMO) communication networks. Our approach tackles a multi-objective optimization problem aiming to maximize network utility while meeting stringent quality of service requirements in M-MIMO networks. To address the non-convex and nonlinear nature of this problem, we introduce a novel multi-agent DQN framework. This framework defines a large action space, state space, and reward functions, enabling us to learn a near-optimal policy. Simulation results demonstrate the superiority of our Parameterized Deep DQN (PD-DQN) approach when compared to traditional DQN and RL methods. Specifically, we show that our approach outperforms traditional DQN methods in terms of convergence speed and final performance. Additionally, our approach shows 72.2% and 108.5% improvement over DQN methods and the RL method, respectively, in handling large-scale multi-agent problems in M-MIMO networks.

This research aims to investigate the complex relationship between logistics optimization and sales performance among Small and Medium Enterprises (SMEs) operating on Shopee Indonesia. The research method uses a mixed methods approach, qualitative insights from interviews with SMEs provide an in-depth understanding of the challenges and positive impacts associated with logistics optimization. Qualitative insights from interviews highlight the challenges faced by SMEs, including financial constraints and lack of awareness, while also highlighting the positive impact of efficient logistics on customer satisfaction, loyalty and sales growth. Quantitative analysis strengthens these qualitative findings through statistically significant correlations. The adoption percentage in the area of logistics optimization indicates a proactive approach among SMEs in simplifying their operations. Financial impact data reflects the mixed results experienced by SMEs, with a significant percentage reporting significant improvements in profitability. The integration of qualitative and quantitative insights provides a holistic understanding of the challenges, strategies and outcomes associated with logistics optimization.
 Keywords: Sales Performance, Logistics Optimization, Shopee Indonesia

Previous work was done to compare the static resource allocation (StRA) scheme and scalable resource allocation (ScRA) scheme within an integrated GSM and GPRS network (see Tang, S. and Tafazolli, R., Vehicular Technology Conference - Spring, 2003). In order to maintain the QoS for both, the ScRA scheme assumed that the load history database was a credible estimate of the present state of traffic intensity in the cell area. In cases when there is a sudden change in traffic conditions, the ScRA scheme may not be able to perform well. Hence, in order to cope with sudden changes of traffic in the network, we propose to incorporate prediction schemes into the ScRA algorithm. For comparison purposes, we present and evaluate three prediction strategies, namely linear regression prediction, exponential regression prediction, and current value prediction, in six various traffic conditions. In addition, resource utilisation gains between the StRA scheme and the proposed ScRA schemes are also studied. The findings show that ScRA exponential regression prediction scheme can cope best in all the tested traffic scenarios and experiences an average of 0.5 percent of resource utilisation gain over the StRA scheme.

This study investigates the factors affecting estimation accuracy of queue length at signalized intersections under low penetration of connected vehicles. A shockwave-based algorithm is proposed to estimate the maximum queue length and residual queue on a cycle-by-cycle basis. Simulation data collected from three consecutive signalized intersections were used to extract trajectories of connected vehicles under five different market penetration rates and two different traffic conditions (under-saturated and moderate). The results confirm that the queue length estimation process is probabilistic and affected by the stochastic changes in traffic conditions. This probabilistic nature is defined by a queue formation coverage index (QI) that proved to significantly affect the queue length estimation accuracy. Overall, the results show that the accuracy of the queue estimates is acceptable when a QI value of at least 50% is achieved. In such limited data environments, the QI showed potential to help as an assessment tool to evaluate the obtained queue estimates.

Emergency response time often makes difference between life and death in cases that man-made or natural disasters happen. In this paper, we propose a real-time route guidance strategy to reduce the traveling time of emergency vehicles. Our strategy consists of path planning and reverse path planning for emergency vehicles. Since the traffic condition changes at any time, we use current and historical traffic information to plan a path. Furthermore, we propose a lane reservation scheme for emergency vehicles to minimize the crash risks. Before an emergency vehicle arrives at each road segment, Road Side Units (RSUs) will at suitable time start to disseminate the warning message so that other drivers have enough time to take proper actions such as speed deceleration or changing lane. Simulation results show that the proposed approach can reduce more than 20% traveling time of emergency vehicles comparing to the shortest path approach with sirens and lights to alert other drivers. By reducing the traveling time, the emergency rescue teams have higher probability to rescue lives and property in time.