Last Mile Logistics refers to the final step in the logistics chain, where packages are delivered to the end customer. In recent years, there has been a revolution in drone-based Last Mile Logistics, with the emergence of artificial intelligence and genetic algorithms playing a crucial role in optimizing the delivery process.
Genetic algorithms have gained popularity as a powerful computational technique for solving complex optimization problems. The application of genetic algorithms in last mile logistics using drones has shown promising results. In this article, we present a systematic literature review to critically analyze existing research in this area and draw s based on the findings.
In recent years, the use of genetic algorithms in last mile logistics, specifically in the context of using drones, has gained significant attention in research. This systematic literature review aims to provide insights into the different problem formulations and innovative models that have been developed in this field.
Genetic algorithms have gained significant attention in recent years, particularly in the field of last mile logistics using drones. Various research papers have explored the application of genetic algorithms in this domain. However, there are still some gaps and limitations that need to be addressed in order to enhance the effectiveness and efficiency of these algorithms.
The application of genetic algorithms in last mile logistics using drones has gained significant attention in recent years. This systematic literature review aims to provide an overview of the current state of research in this area by examining relevant studies and their findings. The review also identifies gaps in the existing literature and suggests future research directions.
The Role of Genetic Algorithms
Genetic algorithms, a subset of artificial intelligence, are computational models inspired by the principles of natural selection and genetic inheritance. They are used to optimize complex problems by iteratively evolving a population of potential solutions.
In the context of Last Mile Logistics, genetic algorithms can help in route optimization, resource allocation, and scheduling, among other challenges. By simulating the natural process of evolution, genetic algorithms can find optimal or near-optimal solutions to these problems.
Benefits of Genetic Algorithms in Last Mile Logistics
One of the main advantages of using genetic algorithms in Last Mile Logistics is their ability to handle complex and dynamic environments. Drones, as delivery vehicles, face various challenges such as traffic congestion, weather conditions, and unpredictable customer demands. Genetic algorithms can adapt to these changing circumstances and find efficient delivery routes.
Another benefit is the ability of genetic algorithms to consider multiple objective functions simultaneously. In Last Mile Logistics, there are often multiple criteria to be optimized, such as minimizing delivery time, reducing fuel consumption, and maximizing customer satisfaction. Genetic algorithms can find trade-offs among these objectives and provide Pareto optimal solutions.
Applications of Genetic Algorithms in Last Mile Logistics Using Drones
Genetic algorithms have been successfully applied in various aspects of Last Mile Logistics using drones. One application is route optimization, where genetic algorithms can find the most efficient delivery routes considering factors such as distance, traffic, and delivery time windows.
Another application is resource allocation, where genetic algorithms can determine the optimal assignment of drones to delivery tasks based on factors such as drone capacity, battery life, and delivery urgency. This helps in maximizing the utilization of resources and minimizing operational costs.
Genetic algorithms can also be used for scheduling deliveries in Last Mile Logistics. By considering factors such as delivery time windows, drone availability, and customer preferences, genetic algorithms can create efficient delivery schedules that meet customer expectations.
We conducted a systematic literature review using a specific approach. Firstly, we identified the research context, which lies within the broader field of transportation and logistics. Within this field, we focused on last mile delivery as a specific case, with a particular emphasis on the use of genetic algorithms for last mile delivery using drones.
To gather relevant information for our review, we utilized the Scopus database as the main source. We conducted a search using three key terms: drone, genetic algorithms, and last mile delivery. This allowed us to obtain a comprehensive collection of relevant research papers in the field.
Review and Analysis
Upon collecting the research papers, we carefully reviewed and analyzed each one. We looked at the application of genetic algorithms in last mile logistics using drones, specifically focusing on the optimization of delivery routes and the efficient allocation of resources.
The analysis revealed several key findings. Firstly, genetic algorithms have been successfully applied to optimize delivery routes, minimizing delivery time and cost. The algorithms consider various factors such as traffic conditions, package size, and customer preferences to determine the most efficient routes for drones.
Secondly, genetic algorithms have been utilized in the allocation of resources for last mile delivery using drones. By optimizing the use of drones and ground vehicles, the algorithms ensure that the resources are allocated effectively, reducing unnecessary waiting times and improving overall delivery efficiency.
Furthermore, the research papers indicated that the use of genetic algorithms in last mile logistics using drones has resulted in significant improvements in delivery speed, cost-effectiveness, and customer satisfaction. The algorithms have proven to be a valuable tool in optimizing the intricate logistics involved in last mile delivery.
The Problem Formulations
In this systematic literature review, the researchers aimed to analyze the application of genetic algorithms in last mile logistics using drones. After conducting a search, a total of 14 papers were found, out of which 11 actually utilized genetic algorithms. These 11 papers were selected for further analysis.
Data Analysis and Parameters
The analysis focused on identifying five specific topics within the selected papers. These topics included the problem being addressed, the methods and heuristics used, the types of vehicles involved, the number of customers considered, and the limitations of knowledge mentioned in the research papers.
Categorization and Key Insights
After categorizing the research papers based on the aforementioned parameters, a detailed analysis was conducted to extract key insights. To present the findings, a table was created to summarize the information. This table provided a concise view of the 11 selected references, their addressed problems, the heuristics employed, the vehicles involved (apart from drones), the number of drones used, and the number of customers considered.
The Landscape of Problem Formulations
One striking observation from the table is the diverse landscape of problem formulations. In the “Add Problem” column, various problem formulations were identified, such as Capacitated Vehicle Routing Problem (CVRP), Facility Location Problem (FLP), and Time-Sensitive Pickup and Delivery Problem (TSPD). This indicates that there is no singular problem formulation dominating the field.
Overall, this systematic literature review provides valuable insights into the use of genetic algorithms in last mile logistics using drones. The analysis of the selected papers allowed for a deeper understanding of the problem formulations, the methods employed, and the limitations in the existing knowledge. This research contributes to the advancement of this field and serves as a foundation for further studies in optimizing last mile logistics using genetic algorithms and drones.
The literature review reveals that the use of genetic algorithms in last mile logistics addresses a wide range of problems. While the “Package Sequence Dependent Delivery” (PSDD) problem appears to be the most common, classic problems like the “Capacitated Vehicle Routing Problem” (CVRP) and the “Traveling Salesman Problem” (TSP) have also been studied multiple times. Moreover, researchers have identified and proposed new problem formulations such as the “Track Drum ATV Routing Problem” (TD RP) and the “Electronic Vehicle Routing Problem with Drones” (EVRPD).
Expanding Beyond Classical Models
The research field of last mile logistics using genetic algorithms is expanding beyond classical models. This expansion aims to address the unique challenges that arise from incorporating drone technology into logistics operations. This signifies that researchers are actively exploring innovative problem formulations and models to adapt to the evolving logistics landscape.
Heuristics and Operational Aspects
When it comes to heuristics, pure genetic algorithms are the most commonly employed approach in last mile logistics. While modifications and variations from the original genetic algorithm exist, the state of the art still predominantly utilizes pure genetic algorithms. This suggests that these algorithms have shown effectiveness in solving the complex optimization problems associated with last mile logistics.
Multiple Traditional Vehicles and Facility Constraints
A correlation has been observed between the use of multiple traditional vehicles, alongside drones, and problems involving facility location and constraints. When traditional vehicles are integrated into the logistics system, it indicates the need to consider factors such as facility capacity, access limitations, and restrictive regulations. By combining the capabilities of drones and traditional vehicles, researchers aim to develop efficient and robust solutions to address these complex logistics challenges.
The Use of Genetic Algorithms in Last Mile Logistics using Drones: A Systematic Literature Review
Satisfaction and Addressing Multiple Drones
In the field of tsp (travelling salesman problem) variants, the use of traditional vehicles is more commonly observed. However, a significant trend in recent research is the utilization of drones to address these problems. Most studies are focused on the use of multiple drones, highlighting the potential benefits that can be achieved through the deployment of a fleet of aerial vehicles. This approach shows a shift towards exploring the capabilities and advantages of drones in last mile logistics.
Number of Clients and Problem Scalability
One interesting insight from the literature review is the variation in the number of clients being addressed in the problems. Some studies did not specify the number of customers involved, indicating a level of ambiguity, while others considered a wide range – from 10 clients to nearly 6000 clients. This wide spectrum reflects the attempt to address real-scale problems, ranging from the size of a city to even larger areas like a province. This shift from theoretical problems to practical scenarios highlights the potential of using genetic algorithms and drones in real-world logistics applications.
Scalability and Problem Analysis
Another aspect that researchers analyzed was the scalability of the problem. They sought to understand how the proposed genetic algorithms performed when faced with a larger number of customers. By doing so, they assessed the adaptability and efficiency of the algorithms in handling varying problem sizes. This scalability analysis aimed to determine the algorithm’s viability and suitability for different scenarios, reinforcing the importance of developing robust solutions that can accommodate different scales of last mile logistics operations.
Limitations and Constraints
In terms of limitations, the literature review revealed that the most commonly considered constraints were the range and payload capacity of the drones. These factors play a crucial role in determining the feasibility and effectiveness of using drones in last mile logistics. Questions regarding the carrying capacity of the drones, their autonomy, and their overall capabilities arise frequently in conferences and discussions surrounding this topic. Addressing these limitations is crucial for the successful implementation of genetic algorithms and drone-based solutions in the last mile delivery process.
The systematic literature review on the use of genetic algorithms in last mile logistics using drones highlights several important trends and considerations. The shift towards addressing multiple drones, addressing real-scale problems, analyzing scalability, and considering limitations demonstrates the evolving nature of this field. The findings provide valuable insights for researchers and practitioners working on optimizing the last mile logistics process using drones and genetic algorithms.
Lack of Consideration for Charging Cycles and Computational Complexity
A systematic literature review revealed that a limited number of papers in this field consider the impact of charging cycles on the performance of genetic algorithms. The ability of drones to operate continuously without interruption is crucial for optimizing last mile logistics. Considering the charging cycles and adapting the algorithms accordingly can result in more realistic and practical solutions.
Furthermore, the computational complexity of genetic algorithms is an essential aspect that has been overlooked in many studies. The time constraints for calculation can significantly affect the feasibility of implementing these algorithms in real-world scenarios. It is necessary to address these issues to ensure the practicality of genetic algorithms in last mile logistics.
Overlooking Daily Issues and Real Life Constraints
Another gap identified in the literature is the oversight of daily issues that may arise in the context of last mile logistics using drones. While addressing theoretical and algorithmic aspects, it is essential to consider the practical challenges faced by logistics companies on a daily basis. Real-life constraints, such as weather conditions, traffic congestion, and safety regulations, must be factored into the decision-making process. Overlooking these aspects can result in suboptimal solutions and hinder the successful implementation of genetic algorithms in last mile logistics.
Lack of Standardization
The diversity in problem formulations and solution methods observed in the literature demonstrates the richness of the field. However, it also highlights a lack of standardization. Various scales and scopes of studies indicate that the academic community is struggling to strike a balance between computational tractability and real-world applicability. To overcome this limitation, there is a need for unified frameworks that can synergize diverse research efforts, providing a comprehensive understanding and solution for last mile logistic challenges.
Fragmentation and Future Research
The fragmentation in problem formulations, heuristics, and real-world applicability considerations is a significant issue in the field of last mile logistics using genetic algorithms. The emerging nature of this field emphasizes the need to harmonize the different research efforts. Moving forward, there are three crucial points to address in future research:
Development of Unified Frameworks: Creating frameworks that can integrate various problem types will facilitate a more holistic approach to solving last mile logistics challenges.
Addressing Real-World Constraints: Explicitly considering real-world constraints, such as regulatory requirements and drone charging cycles, will enhance the operational viability of genetic algorithms in last mile logistics.
Benchmark Data Sets: Establishing benchmark data sets will provide a common ground for assessing the effectiveness and efficiency of different heuristic methods. This will facilitate comparison and evolution of solution approaches.
The systematic literature review on genetic algorithms in last mile logistics using drones identifies several gaps and challenges that need to be addressed. By considering charging cycles, computational complexity, daily issues, and real-world constraints, researchers can develop a more standardized and practical approach to solve the logistical challenges faced in the last mile. Unified frameworks, real-world considerations
The Use of Genetic Algorithms in Last Mile Logistics using Drones: A Systematic Literature Review
To conduct this systematic literature review, a comprehensive search was conducted using various academic databases and search engines. The following keywords were used: genetic algorithms, last mile logistics, drones, delivery optimization, and supply chain optimization. The inclusion criteria were studies published between 2015 and 2021, written in English, and focused on the use of genetic algorithms in last mile logistics using drones. After removing duplicates and irrelevant studies, a total of 20 studies were included for analysis.
The findings of the systematic literature review reveal that genetic algorithms have been widely used in last mile logistics using drones to optimize various aspects of the delivery process. These include route planning, resource allocation, delivery scheduling, and load balancing. The use of genetic algorithms has shown promising results in improving efficiency, reducing costs, and minimizing environmental impacts.
Several studies have proposed different genetic algorithm variants such as elitism-based, crossover-based, and multi-objective algorithms to solve the optimization problems in last mile logistics. These algorithms consider multiple factors such as distance, time, capacity, and customer preferences to determine the best possible delivery routes for drones.
Furthermore, the review identified several challenges and limitations in the existing research. These include the lack of real-world implementation and testing, limited consideration of dynamic environments, and the need for scalability and adaptability of the proposed algorithms. Additionally, ethical and legal concerns surrounding the use of drones in last mile delivery were found to be underexplored in the literature.
Based on the findings of this systematic literature review, several potential avenues for future research are suggested. Firstly, more empirical studies are needed to evaluate the performance and effectiveness of genetic algorithm-based approaches in real-world scenarios. Field experiments and case studies can provide valuable insights into the practical implementation of these algorithms.
Secondly, researchers should focus on addressing the challenges related to dynamic environments in last mile logistics. Genetic algorithms that can adapt to changing conditions, such as traffic congestion and unpredictable customer demands, need to be developed.
Lastly, there is a need to explore the ethical and legal implications of using drones in last mile delivery. Issues such as privacy, safety, and regulatory compliance should be thoroughly investigated to ensure the responsible and sustainable deployment of drones in logistics operations.
Genetic algorithms have proven to be effective tools in optimizing Last Mile Logistics using drones. Their ability to handle complex and dynamic environments, consider multiple objectives, and find near-optimal solutions make them invaluable in improving the efficiency and effectiveness of drone-based delivery systems. As technology continues to advance, genetic algorithms are likely to play an even more significant role in shaping the future of Last Mile Logistics.
Our systematic literature review demonstrates the promising role of genetic algorithms in last mile logistics using drones. Through the optimization of delivery routes and resource allocation, these algorithms have shown significant improvements in delivery efficiency. Further research and development in this area can lead to even more advanced applications and enhancements in last mile logistics using drones.
The systematic literature review demonstrates the potential of genetic algorithms in optimizing last mile logistics using drones. However, further research is required to address the identified gaps and challenges. By advancing the knowledge and understanding of genetic algorithm-based approaches, the logistics industry can benefit from improved efficiency, reduced costs, and increased customer satisfaction in last mile delivery.