Application of Unmanned Aerial Vehicles in Logistics: A Literature Review

Abstract

The booming development of e-commerce has brought many challenges to the logistics industry. To ensure the sustainability of the logistics industry, the impact of environmental and social sustainability factors on logistics development needs to be considered. Unmanned Aerial Vehicles (UAVs)/drones are used in the logistics field because of their flexibility, low cost, environmental protection and energy-saving advantages, which can achieve both economic benefits and social benefits. This paper reviews 36 studies on UAVs applications in logistics from the Web of Science database from the past two years (2021–2022). The selected literature is classified into theoretical models (the traveling salesman problem and other path planning problems), application scenarios (medical safety applications and last-mile delivery problems) and other problems (UAV implementation obstacles, costs, pricing, etc.). Finally, future directions of UAVs are proposed, such as different application scenarios that can be considered and different algorithms that can be combined to optimize paths for UAVs to specific flight environments.

1. Introduction

The widespread adoption of green vehicles in logistics may help alleviate problems such as environmental pollution, global warming and oil dependency. In this context, Unmanned Aerial Vehicles (UAVs)/drones may represent a useful and innovative means of transportation [1]. UAVs are unmanned aircrafts that use radio remote control equipment and self-contained program control devices or are operated completely or intermittently by on-board computers.

As a technology with high flexibility, low cost, environmental protection, energy conservation and other advantages, UAVs are applied in logistics. The application of UAVs can reduce the number of vans or lorries circulating in the city, thereby reducing traffic congestion, noise pollution and pollution emissions. The application of UAVs in the field of logistics can be subclassified into: regional UAV transportation, UAVs express delivery (terminal distribution), UAVs rescue (emergency logistics) and UAVs storage management (inventory, inspection, etc.), among which regional UAV transportation and UAV terminal distribution are the main forms. There are two types of UAV distribution systems, of which the main forms are branch line UAV transportation and UAV terminal distribution. The UAVs used by two well-known logistics companies in China serve as an example: one is the “point-to-multiple” UAV logistics distribution mode represented by JD, which involves control of multiple UAVs, while the other is the “peer-to-peer” UAV logistics distribution mode represented by S.F. Express, which involves control of a single UAV.

Review articles on UAVs in logistics from 2020 to 2022 are summarized in Table 1. Of these articles, some [2,3,4] only studied the path planning problem of UAVs in the logistics field under specific backgrounds. One study [5] focused on the concepts and decision-making issues related to the last-mile delivery, and other studies [6,7] mainly considered the truck-UAV collaborative delivery problem. There are currently no reviews summarizing the literature related to the application of UAVs in logistics in 2021–2022.

Different mathematical models and research algorithms are reviewed considering some bottlenecks in the technical and operational management of UAVs, such as intelligent obstacle avoidance, flight distance, endurance and load capacity. We considered review articles on drones in logistics published between 2020 and 2022 (see Table 1), mainly summarizing the literature before 2020. Three review articles [3,4,7] included literature from 2021 to 2022, but only six papers were covered. Some important research advances have been made regarding drones in logistics from 2021 to 2022, but there are no corresponding review articles yet. Thus, this paper reviews 36 studies in the Web of Science database, summarizes the current research status, finds research deficiencies and then explores the future research directions in this field.

2. Methodology

In this paper, we review the literature on drones in logistics from 2021 to 2022 and summarize the content of the research related to logistics drones in the last two years. The purpose of this review is as follows:

• To present a framework of the content of the last two years of drones in logistics.

• To review the literature on logistics drones in the last two years and classify these studies into two main categories: theoretical models and application scenarios. In addition, we explore the relationship between the use and sustainability of drones in theoretical models and application scenarios.

• Current research priorities and hotspots are identified and research gaps in the applied branches of logistics drones are understood for future research.

2.1. Literature Collection

The literature collection and selection design is shown in Figure 1. The Web of Science database was initially searched for the terms “drone AND logistics” and “UAV AND logistics”. The year of publication was limited to 2021 and 2022, the document type was limited to papers and the language of the literature was restricted to English. As of October 2022, 289 articles are retrieved. Then, we retained the results from business, operational research and management science journals, and 87 papers were obtained. Finally, 36 studies were selected according to the literature content which must be relevant to logistics. The literature screening criteria are shown in Table 2.

2.2. Literature Collection Results

After the survey of the literature, the largest number of studies was collected from the European Journal of Operational Research and Transportation Research Part E: Logistics and Transportation Review. The journals used in the literature review are shown in Table 3.

3. Application of UAVs in Logistics

From the contents of Table 2, it can be seen that most previous reviews [2,3,4,7] focused on summarizing the theoretical models, while some [5,6] focused on summarizing the application scenarios. In this paper, we summarize the literature on logistics drones in the last two years based on the theoretical models and application scenarios and propose the following content framework (see Figure 2).

3.1. Theoretical Models

As can be seen from Table 2, the previous reviews [2,3,4,7] focused on summarizing the route planning problem in theoretical models, mainly involving the optimization of paths, which specifically include the mathematical models, the solution methods and the types of UAVs used in the literature. Due to the complexity and dynamics of the path environment, it is not easy for UAVs to accomplish distribution tasks under various safety risks (e.g., crashes and collisions), and the development of efficient and flexible UAV path planning algorithms have become inevitable [8]. The relevant literature in the last two years also focuses on the study of route planning problems in the theoretical model. Many scholars used different algorithms to achieve the goals of path minimization, cost minimization, etc. Among them, there is much literature on the traveling salesman problem. Considering that the traveling salesman problem belongs to a special path planning problem and partially overlaps with the research content of the last mile problem, the traveling salesman problem is classified into one category and other path planning problems are classified into another category.

3.1.1. The Traveling Salesman Problem

The traveling salesman problem is a special type of path planning problem. A path planning problem involves finding an optimal path from the starting point to the destination under the given constraints (avoiding obstacles, minimizing energy consumption, shortest path, least time) by using different means of transportations, and delivering the designated goods to the destination safely and accurately. The traveling salesman problem is a typical combinatorial optimization problem. Algorithms commonly used to solve this problem include the ant colony optimization algorithm, seagull algorithm, differential evolution algorithm, sparrow search algorithm and many others.

As the research contents of the traveling salesman problem overlap with the path planning problem and the last-mile problem and considering that there are many studies on the traveling salesman problem, the traveling salesman problem is placed in its category, and relevant literature is shown in Table 4.

Considering the limitation of delivery time, under the premise of setting a time window, Mauro Dell’Amico et al. proposed that trucks or UAVs complete customer delivery services, using a branch and bound algorithm especially designed to efficiently target small instances up to 15 customers, in order to minimize the time required to serve all devices [9]. Juan C. Pina-Pardo et al. introduced the traveling salesman problem with release dates and UAVs resupply, which consists of finding a minimum time route for a single truck that can receive newly available orders an route via a UAVs sent from the depot, and they developed a Mixed-Integer Linear Program and a solution approach for larger instances. The experiments show that using UAVs for resupply can reduce the total delivery time by up to 20% [11].

In addition to setting a time limit, the problem of UAVs load limit is also considered when using the UAVs, Michael Dienstknecht et al. researched the problem of UAV resupplying the remaining goods after the truck loads a part of the goods, in order to minimize the total delivery cost, developed suitable optimization approaches and applied them in static and dynamic problem settings [10].

Due to the actual application of the traveling salesman problem, UAV distribution routes need to access different and varied environmental factors. Justo Puerto and Carlos Valverde model the design of routes of UAVs and other vehicles that must visit a number of geographical elements to deliver goods or services and present two families of mathematical programming formulations. The results show that these models are useful and these formulations can solve optimality medium size instances of sizes similar to other combinatorial problems [12]. Nathalie Grangeon et al. used a mixed meta-heuristic algorithm, a mixed integer linear programming formula and a simple branch-and-cut method to study the problem of using UAVs to transport goods to urban areas [13].

3.1.2. Other Path Planning Problems

Aside from the traveling salesman problem, this paper classifies other path planning problems into one category. The relevant literature is shown in Table 5.

Under the time window restriction, Felix Tamke and Udo Buscher developed a new mixed integer linear programming (MILP) model for the vehicle routing problem with UAVs (VRPD) with two different time-oriented objective functions for vehicle routing to maximize the total amount of goods distributed. The results show that integrating truck UAVs tandems into transportation systems not only can lead to improvements regarding the speed of deliveries but also can be used to reduce fleet size [14]. Chen Cheng et al. developed an adaptive large neighborhood search heuristic algorithm to deal with the VRPTWDR problem; the results show that using robots for customers decreases the objective value dramatically [17]. Dyutimoy Nirupam Das et al. proposed a novel mechanism that synchronizes UAVs and delivery trucks, and developed a multi-objective optimization model to minimize the travel costs and maximize the customer service level in terms of timely deliveries [19]. R. J. Kuo et al. explored the cooperation of trucks and UAVs by developing a model for the vehicles routing problem with UAVs that considers the presence of customer time windows. A mixed-integer programming model is presented to minimize the total traveling costs [21]. Luigi Di Puglia Pugliese et al. focused on the energy consumption of the UAVs that we assume to be influenced by atmospheric events; they propose a decomposition approach based on Benders combinatorial cuts, so as to prevent energy interruption in the worst case [26]. Adriano Masone et al., assuming that a UAV is capable of carrying multiple packages at a time and that can be launched and retrieved along an edge, as well as a flexible launch/retrieval site set, developed a lower bound for the problem and an algorithm that improves the existing heuristic algorithm; the results suggest that the ability to launch along an edge has a nontrivial impact on objective values on truck-and-UAV coordination problems [20].

In the path planning problem, the detailed setting of routes and stations is also one of the keys to obtain the best path. Abhishake Kundu et al. presented a routing heuristic for the Flying Sidekick Traveling Salesman Problem. They developed a novel split algorithm that utilizes the shortest path approach for determining the optimal routing solution to a given order of customer locations [15]. Mohamed R. Salama and Sharan Srinivas pointed out that the existing truck-UAV tandems predominantly restrict the UAVs launch and recovery operations to customer locations. In order to solve these problems, they introduced a new variant of truck-UAV tandem that allows the truck to stop at non-customer locations (referred to as flexible sites) for UAVs [24].

In the process of achieving the goal of path optimization, the characteristics of UAVs should also be considered. Munjeong Kang and Chungmok Lee developed an exact algorithm based on the logic-based Benders decomposition approach, with the goal of minimizing the total sum of truck travel and waiting times for UAVs to return after deliveries [16]. Yang Xia et al. considered the social and environmental problems of the UAVs, such as the impact of battery wear and waste. In order to maintain economic performance and minimize negative environmental impacts, the team-sharing solution supporting the blockchain is discussed to optimize the operation of the UAV, so as to achieve multiple goals of the shortest path and the least charging time [18]. Jose Miguel Leon Blanco et al. have developed a new agent-based method to solve the logistics problem of truck and multi-UAV teams under the constraints of UAV battery capacity, truck number and so on, so that the UAV and truck can complete the distribution task in the shortest time [22]. Considering the UAV’s flight range, carrying capacity and demand distribution, Yichen Lu et al. applied the truck and UAVs’ collaborative distribution mode to humanitarian logistics and established a multi-objective mixed integer linear programming (MILP) model with two objectives. Finally, based on actual cases, they investigated the impact of different UAV parameters on the efficiency of vehicle aircraft collaborative distribution [23].

Cl’ement Lemardel’ et al. provided strategic insights into the complex environment of logistics operation in the last-mile deliveries of the city, estimated the operation cost using continuous approximate equations, applied the developed mathematical formulas to two different examples—namely, a Paris suburb and Barcelona—and finally conducted a sensitivity analysis on the model [25].

3.1.3. Summary of Theoretical Models and Related Sustainability Issues

The main purpose of path planning is to achieve path optimization, but the more vehicles in circulation and the longer the distance traveled by vehicles, the more energy consumption and the greater the environmental impact associated with the delivery phase. Currently, most drones are battery-powered. Compared to traditional transportation methods, drones are more environmentally friendly with fewer carbon emissions and energy consumption during the delivery process [27]. The increased delivery of goods may increase road traffic congestion, while drones working in the sky, controlled only by the operator through a remote network, can effectively avoid congestion problems, not only saving time but also shortening the transport distance, thus reducing energy consumption [28].

3.2. Application Scenarios

A previous review [5] summarized the novel delivery concepts and decision problems for last-mile delivery in application scenarios. Another review [6] focused on summarizing the latest optimization methods and synchronization between trucks and drones. The literature related to logistics drones in the last two years mainly focuses on last-mile delivery problems and medical safety issues in application scenarios. With vehicle congestion and customer dispersion, timely delivery has become a challenge for last-mile delivery. In recent years in the environment of the COVID-19 epidemic, the rapid spread of viruses and the high rate of infection in the population have made logistics delivery more difficult. The safety of drone delivery in medical safety has become one of the research topics for many scholars. In addition, the pricing and cost of UAVs have also received a lot of attention.

3.2.1. Medical Security Issues

A UAV has a strong ability of delivery, enabling it to deliver and carry medicines and other disaster relief supplies to designated locations. UAVs in the medical supply chain can deliver life-saving essential drugs or blood to places in urgent need in a very short time, overcoming the difficulty of timely delivery by ordinary distribution methods. In addition, the UAVs reduce the labor cost in this process, and it can effectively avoid infection. For example, in the process of epidemic prevention, UAVs can be used to spray disinfectants, with better disinfection and sterilization effects than traditional methods. Moreover, the use of UAVs also avoids interaction between the delivery person and the customer, thus reducing the possibility of close contact between people. The relevant UAV’s medical safety-related literature is summarized in Table 6.

UAV rescue is a special application of UAVs in medical safety, which has not been widely used. Patrick Holzmann et al. conducted a behavioral study investigating the relationship between personal attitudes, perceptions and intentions for UAV adoption in the context of mountaineering and related activities involving medical accidents. Original survey data of 146 mountain rescuers were analyzed using moderated OLS regression analysis. Results indicate that the behavioral intention to use UAVs in mountain rescue missions is driven by the expected performance gains and facilitating conditions [29]. Idris Jeelani and Masoud Gheisari examined UAVs integration in construction workplaces from a health and safety perspective, categorized the potential ways UAVs might affect the construction workers’ health and safety and used reasoning and VR visualization technology to identify the risks brought by UAVs, so as to improve the understanding of human–computer interactions, and developed regulatory and technical measures to ensure the safety of UAVs in construction [32].

As time control is critical when UAVs are involved in medical safety incidents, Guowei Zhang has developed a collaborative truck-and-UAV system as a post-disaster assessment tool for use by humanitarian relief networks. The proposed system comprises a UAV equipped with a camera that can launch from a truck to collect information from both nodes and links of a post-disaster transportation network. This study was the first to consider the problem of collaborative truck-and-UAV routing optimization with the goal of profit maximization [30]. Zabih Ghelichi et al. developed an optimization model to optimize the logistics for a fleet of UAVs for the timely delivery of medical items (e.g., medicines, test kits and vaccines) to hard-to-access locations. The UAV starts from the urban provider and visits one or more charging platforms to minimize the total completion time of serving all demand points [31].

The delivery of UAVs is restricted by natural conditions such as geography and weather. Maximilian Kunovjanek and Christian Wankmaüller investigated how UAVs can be used to distribute viral tests to potentially infected patients. The private entities are retrofitted for the distribution of essential goods in the case of emergency and potential performance gains are analyzed through a mathematical time and cost model [33]. Debapriya Banik et al. recognized that since UAVs have varying characteristics, such as flight distance, payload-carrying capacity, battery power and so on, selecting an optimal UAV for a particular scenario becomes a major challenge for decision-makers. To address this issue, a decision support model has been developed to select an optimal UAV for two specific scenarios related to medical supplies delivery [34].

3.2.2. Last-Mile Delivery Problems

Last-mile logistics plays an important role in the distribution of goods to end customers. Efficient and effective last-mile logistics is crucial not only for reducing the related time and costs, but also to address sustainability issues related to the environment and road congestion. Last-mile delivery of goods is often carried out by vans, most of which have internal combustion engines. UAVs may represent a helpful and innovative transport system to decrease environmental and noise pollution and congestion [35]. In addition to completing basic transportation or distribution tasks, UAVs must also address other technical, economic, social and other requirements to achieve win-win economic and social benefits. The relevant literature on last-mile delivery is summarized in Table 7.

Combining the natural and self-limiting conditions in the use of UAVs, Luigi Di Puglia Pugliese et al. studied the delivery problem of truck fleets equipped with UAVs to provide services for all customers at the lowest cost under the constraints of the time window, capacity and flight endurance [36]. Emrah Demir et al. studied the latest technological progress and problems to be solved related to the logistics, business and social innovation of the last mile; discussed the UAVs, delivery robots, truck formation, collection and delivery points, collaborative logistics, comprehensive transportation, decarburization and advanced transportation methods; and finally, provided suggestions for strategies that may promote the adoption of new and effective technologies and innovation [38].

Compared with traditional distribution methods, UAV distribution methods are more novel and convenient but have more potential uncertainties. Merkert et al. conducted a survey with 709 interviewees in Australian cities. The purpose of this research was to reveal consumer preferences towards innovative last-mile parcel delivery and more specifically unmanned aerial delivery drones, in comparison to posties (traditional postal delivery) and the recent rise of parcel lockers in Australia [37].

3.2.3. Summary of Application Scenarios Summary and Related Sustainability Topics

In application scenarios, last-mile delivery is one of the trickiest stages of the logistics process, especially in urban environments where the increasing number of packages delivered to customers increases the number of delivery trucks entering the city center, thus increasing congestion, pollution and negative health impacts. Drone delivery methods can reduce the environmental burden associated with last-mile delivery [39]. It is also necessary to consider the impact of parcel delivery speed on user satisfaction, as we desire fast delivery, and compared to traditional parcel delivery methods, drones can deliver directly across time and space. Drones can deliver parcels directly from the cargo center to the doorstep, instead of putting the goods in the transit center before delivery, thus reducing the sum of paths and saving energy and time [40].

In the practice of medical safety, especially during the epidemic of COVID-19, the virus spreads quickly and the infection rate of the population is high. In the face of emergencies, traditional material delivery methods cannot achieve an efficient and timely delivery level, and at the same time, traditional transportation methods require a lot of manpower support, which increases the movement of people and adds risk to the spread of the virus. Drones can replace manual operations and can deliver timely supplies to medical personnel and special groups in special environments, such as nucleic acid testing samples and daily supplies for isolated groups to avoid cross-contamination. For short-distance delivery, drone delivery is usually faster than truck delivery. It highlights the sustainability of drones.

3.3. Other Problems

In addition to the theoretical model and application scenarios, nine studies remain that are not summarized, including studies on UAV scheduling problems, UAV pricing problems and obstacles to UAV implementation. This paper classifies these remaining studies into the “other problems” category, as shown in Table 8.

UAV delivery will encounter more difficulties than manual delivery. Bhawesh Sah et al. identified and prioritized the barriers to UAV logistics implementation based on their criticality by using the fuzzy Delphi method (FDM) and the analytic hierarchy process (AHP). Initially, 34 barriers are identified through expert opinion and an extensive literature review. Finally, the management significance of the research results that can help practitioners and decision-makers to effectively implement drones in the logistics department is discussed [41]. Aditya Kamat et al. analyzed the various barriers hindering the implementation of UAVs in humanitarian logistics for both developed and developing nations, and they propose an interval-valued intuitionistic fuzzy set (IVIFS) to calculate a UAV implementation hindrance index (DIHI) [42].

Due to the limitation of delivery time, Yohei HAZAMA et al. proposed a GA to solve the package delivery scheduling problem. They define the parcel delivery scheduling problem as finding the assignment of customers to both the UAVs and their takeoff points. The purpose is to find the near-optimal solution in a short time to reduce the cost and time required for package delivery [43]. Kai Wang et al. proposed a piggyback transportation problem based on the last-mile flying warehouse; they formulated the Piggyback Transportation Problem, investigated its computational complexity and derived suitable solution procedures [44].

Considering the economic benefits of using UAVs for delivery, Zhi Pei et al. focused on the UAVs sharing system, wherein revenue management becomes vital in terms of pricing, UAV hiring cost and service-related cost. A time-varying and price-sensitive queueing model is formulated, where the customer behaviors are taken into account. Different algorithms are used to solve the problem of high, medium and low service quality objectives [45]. Yaohan Shen et al. studied a multi-warehouse UAV delivery system, considering the allocation rule that all warehouses share the UAVs and the allocation rule that each warehouse owns its UAVs. Both plug-in charge and battery swap strategies are investigated for battery management. They developed a cost minimization model for cost analysis [46]. Marc Antoine Coindreau et al. considered global costs, including fixed daily vehicles fares, driver wages and fuel and electricity consumption to power trucks and UAVs, and proposed a mixed-integer linear programming formulation and an adaptive large neighborhood search to solve a parcel delivery problem with a fleet of trucks embedded with UAVs [47]. Suttinee Sawadsitnag et al. considered the uncertain factors of the implementation of UAVs in the logistics field and proposed the BCoSDD framework, which is composed of three functions: package assignment, shipper cooperation formation and cost management in order to minimize their UAVs delivery cost [48]. Considering the limited capacity of UAVs, James F. Campbell et al. proposed a branch-cut algorithm and a mathematical method to address the Length Constrained K-UAVs Rural Postman Problem (LC K-DRPP) [49].

4. Future Research Directions

The application of UAVs in the field of logistics has largely reduced the workload of staff and improved people’s quality of life. The future of UAVs will still involve extensive further development. This paper addresses the future research directions for the application of UAVs in the field of logistics from several aspects, as shown in Table 9.

4.1. Path Planning Problems

The UAVs path planning problem accounts for the largest proportion of the reviewed literature. The use of algorithms is also very rich, as many scholars focus their UAVs research on the study of route optimization. However, further advancement and breakthroughs are still needed in the UAVs path planning research. The prospects in this area are discussed below.

Since the flight altitude of UAVs for industrial applications is below 400 m in the low-altitude environment, this complex environment presents new requirements for the path planning algorithm for UAVs applications. Therefore, UAV path planning needs to further consider different operation scenarios (altitude, wind speed, temperature, humidity, etc.) and make corresponding decisions for specific flight environments to adapt to different application situations.

For path planning, there are few studies on UAV operation in urban environments. In the future, UAV path planning can focus on improving the comprehensiveness of flight environment construction, the efficiency of path planning algorithms and the cooperative operation efficiency of UAVs and distribution vehicles, or it can combine different algorithms for path planning, which can effectively improve the real-time decision-making ability of the algorithm. Finally, most of the current path planning algorithms for UAVs involve path planning for a single UAV. As the development of UAVs is moving towards clustering and intelligence, future research direction can consider using more comprehensive path planning methods for UAV clusters.

4.2. Medical Safety Issues

Considering the particularity of UAV distribution in the epidemic environment, due to airspace problems, distribution tasks and other constraints, we can establish a multi-constraint medical UAV distribution system model with the minimum flight time, energy consumption and risk as the objective function, enabling us to quickly solve the distribution path. Future research on UAVs can enhance the flexibility of operating UAVs through precise algorithms, and also control the cost of UAV applications in medical safety by designing more reasonable drug-carrying devices. The core technology of drones can also be further solved to improve the efficiency of drones for medical supplies delivery.

4.3. Last-Mile Delivery Problems

Last-mile distribution accounts for more than 30% of the total cost of logistics distribution. Problems such as scattered customers and serious loss of parts greatly increase the cost of distribution. With the development of digitalization, logistics facilities and payment systems have been continuously improved. Solving the distribution of the last mile has become the key to the success of the express industry and e-commerce. The future research direction can combine the path planning of UAVs to seek the shortest path and the most efficient with last-mile delivery to solve the distribution problems of UAVs.

5. Conclusions

Due to the continuous updating and maturation of UAV-related technologies, UAVs continue to show unique performance advantages when applied in logistics, such as low cost, environmental protection and energy savings. The application of UAVs can, thus, reduce the impact on the environment and society and contribute to the sustainability of logistics. This paper reviewed 36 studies from 2021 to 2022 in the Web of Science database. There is little overlap between the studies selected for this paper and the literature selected for the last three years (2020–2022) of UAV-related reviews. The selected studies are classified into theoretical models (the traveling salesman problem and other path planning problems), application scenarios (medical security application and last-mile delivery problem) and other problems (UAV implementation obstacles, costs, pricing, etc.).

Reasonable path planning can ensure a very reasonable vehicle transportation route, which will have a great influence on improving distribution speed, reducing costs and increasing efficiency. Among the theoretical models, previous reviews have focused on the UAV path planning problem and summarized many algorithms that can achieve path optimization, specifically including mathematical models, experimental constraints, UAV types, etc. Due to the importance of path planning in logistics, the literature has also focused on the path problem of UAVs in the past two years. Based on the models and algorithms used by previous scholars, algorithm optimization has been carried out and new solutions have been proposed; many scholars have tested the algorithms through examples to verify the superiority and accuracy of the algorithms, and many of them have studied the traveling salesman problem. We believe that the path planning problem of UAVs remains a hot research topic at present and future research trends.

In the application scenarios, drones have attracted the attention of scholars for their potential to reduce the environmental burden in last-mile delivery [50]. The previous reviews focused on the concepts related to the last-mile delivery problems of UAVs and the decision-making problems, also including the issues of path optimization and implementation barriers in the collaborative delivery of trucks and UAVs. The related literature in the last two years explored people’s preferences for novel last-mile delivery methods as well as investigated the last-mile issues to be addressed based on the emphasis on green logistics and urban logistics concepts. In addition to last-mile delivery problems, we also summarized the problems of drones applied in medical safety. In the epidemic environment, the indicators of logistics delivery are not only to complete basic delivery tasks but also to meet the requirements of epidemic prevention and monitoring to achieve accurate, safe and timely delivery. The drones can replace manual delivery in the air in a contactless way to complete tasks such as spraying disinfectant and delivering drugs.

At present, logistics distribution is mainly concentrated on the ground, and air pollution and traffic congestion are becoming increasingly serious. Most drones are battery-driven; thus, compared to traditional delivery vehicles, drones have fewer carbon emissions and lower energy consumption in the distribution process, and are more environmentally friendly. In consideration of environmental protection, cost, delivery efficiency and other comprehensive factors, from the perspective of sustainability, the application of drones in logistics may be a good choice. With the popularization and improvement of UAV technology, drones are expected to achieve a larger scale of parcel delivery [51].

However, considering that this paper only selected literature from the Web of Science database and only retained literature from business, operational research and management science journals, some literature from other databases or journals may be missing. Moreover, the contents of this review were obtained by the authors through reading and sorting the identified literature. If a literature measurement and other methods are used in the future, different results may be produced.

Author Contributions

Conceptualization, Y.L. and M.L.; methodology, Y.L. and M.L.; validation, M.L.; formal analysis, M.L.; investigation, M.L.; resources, M.L. and D.J.; data curation, M.L.; writing—original draft preparation, M.L.; writing—review and editing, Y.L. and D.J.; visualization, M.L.; supervision, Y.L.; project administration, Y.L. and D.J.; funding acquisition, Y.L. and D.J. All authors have read and agreed to the published version of the manuscript.

Institutional Review Board Statement

Not applicable.

Informed Consent Statement

Not applicable.

Data Availability Statement

Not applicable.

Conflicts of Interest

The authors declare no conflict of interest.

Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Figures and Tables

Figure 1. Research design.

Figure 2. Content framework.

Table 1

Review articles on UAVs in logistics from 2020–2022.

Reference	Main Topic	Theoretical Models	Application Scenarios	The Year of the Reviewed Literature
[2]	This paper researches literature in Elsevier, Wiley, Springer, Scopus, ResearchGate and Google Scholar databases, with a focus on UAV path planning from the perspective of operations research in the context of package delivery.	√		2006–2020
[3]	This paper investigates the established and novel concept of last-mile delivery, emphasizing the decision-making problems to be solved when establishing and operating the concept and summarizing the operational research methods to solve these problems.		√	2000–2020
[4]	This paper researches various optimization problems (mathematical models, solution methods, etc.) and implementation obstacles in the application of UAVs and UAV-truck cooperation.		√	2001–2020
[5]	This paper classifies delivery systems for UAVs and their associated path planning problems.	√		2001–2021
[6]	This paper researches the path planning problem of two-echelon networks with the modeling mechanism of connecting two echelons.	√		2000–2021
[7]	This paper classifies the UAVs routing problem and reviews and describes the popular algorithms widely used in the existing literature.	√		2001–2022
This paper	This present paper reviews research on the application of UAVs in logistics, including theoretical models (the traveling salesman problems and other path planning problems), application scenarios (safety applications and last-mile delivery problems) and other problems (UAV implementation obstacles, costs, pricing, etc.).	√	√	2021–2022

Note: “√” means relevant content is covered in this article.

Table 2

Literature screening criteria.

	Exclusion Criteria	Judgment Method
1	The literature is not about the application of UAVs in logistics (digital technology application, robot application in logistics, UAVs application in other fields, UAVs technology development, communication optimization, etc.)	Reading the title, abstract and keywords of the literature
2	Journal of automation and mathematics, etc.	Reading the journals
3	Other application themes (agriculture, forestry, electricity, meteorology, etc.)	Reading abstract

Table 3

Summary of the journals used in the literature review.

The Journals Used in the Literature Review	Article Number
European Journal of Operational Research	5
Transportation Research Part E: Logistics and Transportation Review	4
Networks	3
International Journal of Physical Distribution Logistics Management	2
Omega-International Journal of Management Science	2
Annals of Operations Research	2
Computers Operations Research	2
Ima Journal of Management Mathematics	2
Expert Systems with Applications	2
IEEE Transactions on Intelligent Transportation Systems	1
Safety Science	1
International Journal of Logistics Management	1
Journal of Optimization Theory and Applications	1
International Journal of Logistics-Research and Applications	1
IEEE Transactions on Vehicular Technology	1
Transportation Science	1
International Journal of Production Research	1
Transport Policy	1
Operations Management Research	1
Journal of Advanced Mechanical Design Systems and Manufacturing	1
Or Spectrum	1

Table 4

Summary of surveyed literature discussing the traveling salesman problem.

Reference	Restrictions	ObjectiveFunction	Approach	Conclusion
[9]	Time window	Minimize the completion time	Branch and bound algorithm, heuristic algorithm	The heuristic algorithm is able to provide state-of-the-art results for medium/large instances.
[10]	Truck passenger capacity	Minimize the completion time	Dynamic programming (DP)	UAV’s resupply can greatly improve delivery performance.
[11]	Time window, UAVs speed and capacity	Minimize the completion time	Mixed integer linear programming	The use of UAVs for resupply can reduce total delivery time by up to 20%.
[12]	UAVs access factors	Minimize cost	Two families of mathematical programming formulations.	Two families of mathematical programming formulations can solve optimization and other combination problems
[13]	The return flight path of UAVs	Minimize the completion time	Mixed meta-heuristic algorithm	The partitioning between vehicles and UAVs would easily be managed, but the completion time objective would certainly be more challenging.

Table 5

Summary of surveyed literature: other path planning problems.

Reference	Restrictions	ObjectiveFunction	Approach	Conclusion
[14]	Time window	Delivery speed and workload	Mixed integer linear programming (MILP) model, branch and cut algorithm	Integrating truck-UAV tandems into transportation systems not only can lead to improvements regarding the speed of deliveries but also can be used to reduce the fleet size without slowing down the delivery process and increasing the workload of truck drivers.
[15]	-	Minimize the completion time	Split algorithm and shortest path method	The algorithm based on the logic-based Benders decomposition approach, which outperforms the state-of-the-art solvers.
[16]	Speed, battery capacity	Minimize the truck path and the waiting time for the return of the UAVs after delivery.	Mixed integer programming formula, the benders decomposition method	Benders decomposition method based on logic has better performance than Cplex.
[17]	Time window	Minimize the completion time	An adaptive large neighborhood search heuristic algorithm	Adaptive large neighborhood search heuristic algorithm can effectively improve the performance of the VRPTWDR problem.
[18]	Vehicles capacity limit, UAVs battery endurance	Find the shortest path and minimize the completion time	Mixed integer programming model, branch and price algorithm	The exact branch-and-price algorithm was developed. Instances of up to 100 customers can be solved optimally by the proposed algorithm.
[19]	Time window	Minimize costs and maximize customer service levels in terms of timely delivery.	Pareto Ant Colony Optimization Algorithm	The collaborative Pareto Ant Colony Optimization algorithm is an efficient solution for parcel delivery.
[20]	Time window, battery life	Minimize the completiontime	Global continuous method	The ability to launch along an edge has a nontrivial impact on objective values on truck-and-UAV coordination problems.
[21]	Time window	Minimize costs	Mixed integer programming (MIP) model, variable neighborhood search (VNS) program	The variable neighborhood search (VNS) program can effectively solve the UAVs vehicle routing problem (VRPTWD).
[22]	UAVs battery capacity, number of trucks and routes	Minimize the completiontime	Iterative greedy algorithm	Agents represent the points that are going to be visited by vehicles. The agent-based system can solve complex optimization problems with high quality.
[23]	Flight range and carrying capacity	Minimize time and maximize the minimum satisfaction rate of demand nodes	Specialized Local Search Operator (HMOEAS) and Ant Colony Algorithm (HACO)	HMOEAS has more advantages than other methods in terms of distribution efficiency of epidemic-resistant materials.
[24]	Routes and Sites	Minimize the completiontime	Mixed Integer Linear Programming Model, Mixed Simulated Annealing	Contrary to existing methods of limiting truck parking locations, the use of flexible sites can greatly improve delivery efficiency.
[25]	-	Minimize costs	Continuous approximation (CA) analysis.	In dense streets, ground autonomous delivery devices have more economic benefits than traditional methods.
[26]	Time window, energy limit	Minimize costs	Benders Combined Cutting Decomposition Method	The algorithm can effectively solve quality problems in terms of cost and energy consumption.

Table 6

Summary of surveyed literature regarding the medical safety issues.

Reference	Restrictions	ObjectiveFunction	Approach	Conclusion
[29]	-	Personal attitude, perception	Moderated ordinary least squares (OLS)	The behavioral intention to use UAVs in mountain rescue missions is driven by the expected performance gains.
[30]	Scheduled time limit	Minimize thetotal route	Mixed integer linear programming (MILP) model	This algorithm can obtain high-quality solutions with a less than 10% optimization gap for all terminal instances within the predefined time limit.
[31]	Timewindow	Minimize thecompletion time	Slot travel scheduling formula	The more UAVs are allowed, the more the system performance depends on the number of charging stations.
[32]	-	Minimize the completion time	-	The noise of UAVs or the presence of UAVs in the working environment may distract workers.
[33]	Weather conditions	-	Mathematical time and cost model	UAVs provide flexible and fast delivery services to deal with powerful disasters.
[34]	Carrying capacity	Select the best UAVs for specific scenarios	matrix approach (GTMA)	UAVs equipped with payload handling capacity flexibility get more preference in urban regions than in other areas.

Table 7

Summary of surveyed literature: the last-mile delivery problem.

Reference	Restrictions	ObjectiveFunction	Approach	Conclusion
[36]	Time window	Minimum cost	Mixed integer linear program	Two UAVs associated with each truck are the best option. Indeed, using three UAVs per truck does not lead to a further reduction in transportation cost.
[37]	-	-	A survey involving stated choice experiments	People prefer traditional postal delivery over UAV delivery, all else equal, but UAV deliveries become competitive with large market shares if they live up to the premise that they can deliver faster and cheaper.
[38]	Capacity, speed	-	The variable domain search algorithm	Customers prefer to have on-time delivery; there is a need for more research focusing on operational and tactical issues related to routing.

Table 8

Summary of surveyed literature regarding other problems.

Reference	Restrictions	ObjectiveFunction	Approach	Conclusion
[41]	-	-	Fuzzy Delphi Method (FDM), Analytic Hierarchy Process (AHP)	Regulations and threats to privacy and security are the most critical barriers to the implementation of UAVs in the logistics sector.
[43]	Fixed route and travel time of truck	Minimize the completion time	Genetic algorithm (GA)	The proposed GA can successfully find an optimal or near-optimal solution faster than an integer programming solver for almost all instances.
[45]	Weather conditions, takeoff/landing time, etc.	Maximize profits	Time-varying and price-sensitive queuing model	The proposed approximation methods not only provide a high-quality joint strategy but also help stabilize the system’s performance.
[46]	Distribution principle	Minimize costs	A neighborhood search algorithm, approximate mean method and heuristic algorithm	When the number of UAVs is not large, and the heuristic can improve the throughput capacity by about 13.31%.
[47]	Time window	Minimize costs	Mixed integer linear programming formula, adaptive large neighborhood search	Truck-and-UAV solutionscan reduce costs by up to 34% compared to traditional truck-only delivery.
[48]	Flight distance, limited time	Minimize costs	Bayesian Shipper Cooperation in Stochastic UAVs Delivery(BCoSDD) framework	This framework can help the shippers plan and schedule their UAV delivery effectively.
[49]	Limit of maximum distance	Minimize costs	Branch cutting algorithm, heuristic algorithm	When the percentage of clients reachable by UAVs is above 50%, expenses related to fuel consumption can be reduced by at least 15%.
[44]	Weight and size	Minimize completiontime	Meta heuristic algorithm, integer programming, greedy algorithm, polynomial algorithm	The authors prove several important properties of our piggyback transportation problem, such as the complexity status, efficient solution algorithms for restricted problem settings and a polynomial-time approximation scheme for the problem.
[42]	-		Graph theory and matrix approach (GTMA) and PERMAN algorithm	Developed and developing countries must first improve their inadequate government regulations regarding humanitarian UAVs.

Table 9

The futures research direction of UAVs in logistics.

	Research Theme	Theoretical Perspective	Approach
Research status	The topic of path research on UAVs is quite varied and the algorithms used are abundant; whether for the traveling salesman problem or the last-mile delivery problem, almost all the relevant literature deals with path optimization to maximize cost savings, with less literature focusing on human–machine interaction, implementation barriers, benefits/costs, etc.	Research on single/multiple UAVs or trucks jointly completing delivery tasks under the constraints of multiple factors such as UAVs endurance and carrying capacity, so as to achieve the goal of saving time and economic costs.	Most of the studies use mathematical models and formulas such as mixed integer linear programming to study the UAV delivery problem under the time window restriction.
Research deficiencies	Most of the current research models are designed for applications in e-commerce and healthcare/emergency services. Other applications, such as food and mail delivery, remain underrepresented in academic discussions.	The literature also lacks comprehensive research on the comprehensive design and operation planning of these delivery modes, and the research on performance comparison of logistics models based on UAVs is also lacking.	The literature research method may omit some literature.
Future research directions	UAVs application planning can further consider different operation scenarios (altitude, wind speed, temperature, humidity, etc.), and conduct research on specific flight environments, combinations of different algorithms, multiple UAVs planning, etc. UAV delivery under dynamic orders can also be studied.	For the use of UAVs in logistics, the contribution and impact of the process of pursuing carbon neutrality can be considered, as well as the use of UAVs in green logistics and green supply chains.	It is possible to improve the accuracy rate and the practical feasibility of experiments when using relevant research methods.

References

1. Patella, S.M.; Grazieschi, G.; Gatta, V.; Marcucci, E.; Carrese, S. The adoption of green vehicles in last mile logistics: A systematic review. Sustainability; 2020; 13, 6. [DOI: https://dx.doi.org/10.3390/su13010006]

2. Macrina, G.; Pugliese, L.D.P.; Guerriero, F.; Laporte, G. Drone-aided routing: A literature review. Transp. Res. Pt. C-Emerg. Technol; 2020; 120, 102762. [DOI: https://dx.doi.org/10.1016/j.trc.2020.102762]

3. Boysen, N.; Fedtke, S.; Schwerdfeger, S. Last-mile delivery concepts: A survey from an operational research perspective. OR Spectr.; 2021; 43, pp. 1-58. [DOI: https://dx.doi.org/10.1007/s00291-020-00607-8]

4. Chung, S.H.; Sah, B. Optimization for drone and drone-truck combined operations: A review of the state of the art and future directions. Comput. Oper. Res; 2020; 123, 105004. [DOI: https://dx.doi.org/10.1016/j.cor.2020.105004]

5. Moshref-Javadi, M.; Winkenbach, M. Applications and Research avenues for drone-based models in logistics: A classification and review. Expert Syst. Appl.; 2021; 177, 114854. [DOI: https://dx.doi.org/10.1016/j.eswa.2021.114854]

6. Li, H.; Chen, J.; Wang, F.; Bai, M. Ground-vehicle and unmanned-aerial-vehicle routing problems from two-echelon scheme perspective: A review. Eur. J. Oper. Res; 2021; 294, pp. 1078-1095. [DOI: https://dx.doi.org/10.1016/j.ejor.2021.02.022]

7. Liang, Y.J.; Luo, Z.X. A Survey of Truck-Drone Routing Problem: Literature Review and Research Prospects. Journal of the Operations Research Society of China; 2022; 10, pp. 343-377. [DOI: https://dx.doi.org/10.1007/s40305-021-00383-4]

8. Li, D.; Yin, W.; Wong, W.E.; Jian, M.; Chau, M. Quality-oriented hybrid path planning based on a* and q-learning for unmanned aerial vehicle. IEEE Access; 2021; 10, pp. 7664-7674. [DOI: https://dx.doi.org/10.1109/ACCESS.2021.3139534]

9. Dell’Amico, M.; Montemanni, R.; Novellani, S. Algorithms based on branch and bound for the flying sidekick traveling salesman problem. Omega-Int. J. Manage. Sci.; 2021; 104, 102493. [DOI: https://dx.doi.org/10.1016/j.omega.2021.102493]

10. Dienstknecht, M.; Boysen, N.; Briskorn, D. The traveling salesman problem with drone resupply. OR Spectr.; 2022; 44, pp. 1045-1086. [DOI: https://dx.doi.org/10.1007/s00291-022-00680-1]

11. Pina-Pardo, J.C.; Silva, D.F.; Smith, A.E. The traveling salesman problem with release dates and drone resupply. Comput. Oper. Res.; 2020; 129, 105170. [DOI: https://dx.doi.org/10.1016/j.cor.2020.105170]

12. Puerto, J.; Valverde, C. Routing for unmanned aerial vehicles: Touring dimensional sets. Eur. J. Oper. Res.; 2022; 298, pp. 118-136. [DOI: https://dx.doi.org/10.1016/j.ejor.2021.06.061]

13. Saleu, R.G.M.; Deroussi, L.; Feillet, D.; Grangeon, N.; Quilliot, A. The parallel drone scheduling problem with multiple drones and vehicles. Eur. J. Oper. Res.; 2022; 300, pp. 571-589. [DOI: https://dx.doi.org/10.1016/j.ejor.2021.08.014]

14. Tamke, F.; Buscher, U. A branch-and-cut algorithm for the vehicle routing problem with drones. Transp. Res. Pt. B-Methodol.; 2021; 144, pp. 174-203. [DOI: https://dx.doi.org/10.1016/j.trb.2020.11.011]

15. Kundu, A.; Escobar, R.G.; Matis, T.I. An efficient routing heuristic for a drone-assisted delivery problem. IMA J. Manag. Math.; 2022; 33, pp. 583-601. [DOI: https://dx.doi.org/10.1093/imaman/dpab039]

16. Kang, M.; Lee, C. An exact algorithm for heterogeneous drone-truck routing problem. Transp. Sci.; 2021; 55, pp. 1088-1112. [DOI: https://dx.doi.org/10.1287/trsc.2021.1055]

17. Chen, C.; Demir, E.; Huang, Y. An adaptive large neighborhood search heuristic for the vehicle routing problem with time windows and delivery robots. Eur. J. Oper. Res.; 2021; 294, pp. 1164-1180. [DOI: https://dx.doi.org/10.1016/j.ejor.2021.02.027]

18. Xia, Y.; Zeng, W.; Xing, X.; Zhan, Y.; Kumar, A. Joint optimisation of drone routing and battery wear for sustainable supply chain development: A mixed-integer programming model based on blockchain-enabled fleet sharing. Ann. Oper. Res.; 2021; pp. 1-39. [DOI: https://dx.doi.org/10.1007/s10479-021-04459-5]

19. Das, D.N.; Sewani, R.; Wang, J.; Tiwari, M.K. Synchronized truck and drone routing in package delivery logistics. IEEE Trans. Intell. Transp. Syst.; 2021; 22, pp. 5772-5782. [DOI: https://dx.doi.org/10.1109/TITS.2020.2992549]

20. Masone, A.; Poikonen, S.; Golden, B.L. The multivisit drone routing problem with edge launches: An iterative approach with discrete and continuous improvements. Networks; 2022; 80, pp. 193-216. [DOI: https://dx.doi.org/10.1002/net.22087]

21. Kuo, R.J.; Lu, S.H.; Lai, P.Y.; Mara, S.T.W. Vehicle Routing Problem with Drones Considering Time Windows. Expert Syst. Appl.; 2022; 191, 116264. [DOI: https://dx.doi.org/10.1016/j.eswa.2021.116264]

22. Leon-Blanco, J.M.; Gonzalez-R, P.L.; Andrade-Pineda, J.L.; Canca, D.; Calle, M. A multi-agent approach to the truck multi-drone routing problem. Expert Syst. Appl.; 2022; 195, 116604. [DOI: https://dx.doi.org/10.1016/j.eswa.2022.116604]

23. Lu, Y.; Yang, C.; Yang, J. A multi-objective humanitarian pickup and delivery vehicle routing problem with drones. Ann. Oper. Res.; 2022; [DOI: https://dx.doi.org/10.1007/s10479-022-04816-y]

24. Salama, M.R.; Srinivas, S. Collaborative truck multi-drone routing and scheduling problem: Package delivery with flexible launch and recovery sites. Transp. Res. Pt. e-Logist. Transp. Rev.; 2022; 164, 102788. [DOI: https://dx.doi.org/10.1016/j.tre.2022.102788]

25. Lemardelé, C.; Estrada, M.; Pagès, L.; Bachofner, M. Potentialities of drones and ground autonomous delivery devices for last-mile logistics. Transp. Res. Pt. e-Logist. Transp. Rev.; 2021; 149, 102325. [DOI: https://dx.doi.org/10.1016/j.tre.2021.102325]

26. Di Puglia Pugliese, L.; Guerriero, F.; Scutellá, M.G. The last-mile delivery process with trucks and drones under uncertain energy consumption. J. Optim. Theory Appl; 2021; 191, pp. 31-67. [DOI: https://dx.doi.org/10.1007/s10957-021-01918-8]

27. Eun, J.; Song, B.D.; Lee, S.; Lim, D.E. Mathematical investigation on the sustainability of UAV logistics. Sustainability; 2019; 11, 5932. [DOI: https://dx.doi.org/10.3390/su11215932]

28. Bruni, M.E.; Khodaparasti, S. A Variable Neighborhood Descent Matheuristic for the Drone Routing Problem with Beehives Sharing. Sustainability; 2022; 14, 9978. [DOI: https://dx.doi.org/10.3390/su14169978]

29. Holzmann, P.; Wankmüller, C.; Globocnik, D.; Schwarz, E.J. Drones to the rescue? exploring rescue workers’ behavioral intention to adopt drones in mountain rescue missions. Int. J. Phys. Distrib. Logist. Manag.; 2021; 51, pp. 381-402. [DOI: https://dx.doi.org/10.1108/IJPDLM-01-2020-0025]

30. Zhang, G.; Zhu, N.; Ma, S.; Xia, J. Humanitarian relief network assessment using collaborative truck-and-drone system. Transp. Res. Pt. e-Logist. Transp. Rev.; 2021; 152, 102417. [DOI: https://dx.doi.org/10.1016/j.tre.2021.102417]

31. Ghelichi, Z.; Gentili, M.; Mirchandani, P.B. Logistics for a fleet of drones for medical item delivery: A case study for louisville, ky. Comput. Oper. Res.; 2021; 135, 105443. [DOI: https://dx.doi.org/10.1016/j.cor.2021.105443]

32. Jeelani, I.; Gheisari, M. Safety challenges of uav integration in construction: Conceptual analysis and future research roadmap. Saf. Sci.; 2021; 144, 105473. [DOI: https://dx.doi.org/10.1016/j.ssci.2021.105473]

33. Kunovjanek, M.; Wankmüller, C. Containing the COVID-19 pandemic with drones-feasibility of a drone enabled back-up transport system. Transp. Policy; 2021; 106, pp. 141-153. [DOI: https://dx.doi.org/10.1016/j.tranpol.2021.03.015]

34. Banik, D.; Hossain, N.U.I.; Govindan, K.; Nur, F.; Babski-Reeves, K. A decision support model for selecting unmanned aerial vehicle for medical supplies: Context of COVID-19 pandemic. Int. J. Logist. Manag.; 2022; [DOI: https://dx.doi.org/10.1108/IJLM-06-2021-0334]

35. Borghetti, F.; Caballini, C.; Carboni, A.; Grossato, G.; Maja, R.; Barabino, B. The Use of Drones for Last-Mile Delivery: A Numerical Case Study in Milan, Italy. Sustainability; 2022; 14, 1766. [DOI: https://dx.doi.org/10.3390/su14031766]

36. Di Puglia Pugliese, L.; Macrina, G.; Guerriero, F. Trucks and drones cooperation in the last-mile delivery process. Networks; 2021; 78, pp. 371-399. [DOI: https://dx.doi.org/10.1002/net.22015]

37. Merkert, R.; Bliemer, M.C.J.; Fayyaz, M. Consumer preferences for innovative and traditional last-mile parcel delivery. Int. J. Phys. Distrib. Logist. Manag.; 2022; 52, pp. 261-284. [DOI: https://dx.doi.org/10.1108/IJPDLM-01-2021-0013]

38. Demir, E.; Syntetos, A.; van Woensel, T. Last mile logistics: Research trends and needs. IMA J. Manag. Math.; 2022; 33, pp. 549-561. [DOI: https://dx.doi.org/10.1093/imaman/dpac006]

39. Li, X.; Gong, L.; Liu, X.; Jiang, F.; Shi, W.; Fan, L.; Xu, J. Solving the last mile problem in logistics: A mobile edge computing and blockchain-based unmanned aerial vehicle delivery system. Concurr. Comput.-Pract. Exp.; 2022; 34, e6068. [DOI: https://dx.doi.org/10.1002/cpe.6068]

40. Munawar, H.S.; Inam, H.; Ullah, F.; Qayyum, S.; Kouzani, A.Z.; Mahmud, M.P. Towards smart healthcare: Uav-based optimized path planning for delivering COVID-19 self-testing kits using cutting edge technologies. Sustainability; 2021; 13, 10426. [DOI: https://dx.doi.org/10.3390/su131810426]

41. Sah, B.; Gupta, R.; Bani-Hani, D. Analysis of barriers to implement drone logistics. Int. J. Logist.-Res. Appl.; 2021; 24, pp. 531-550. [DOI: https://dx.doi.org/10.1080/13675567.2020.1782862]

42. Kamat, A.; Shanker, S.; Barve, A.; Muduli, K.; Mangla, S.K.; Luthra, S. Uncovering interrelationships between barriers to unmanned aerial vehicles in humanitarian logistics. Oper. Manag. Res.; 2022; pp. 1-27. [DOI: https://dx.doi.org/10.1007/s12063-021-00235-7]

43. Hazama, Y.; Iima, H.; Karuno, Y.; Mishima, K. Genetic algorithm for scheduling of parcel delivery by drones. Adv. Mech. Des. Syst. Manuf.; 2021; 15, JAMDSM0069. [DOI: https://dx.doi.org/10.1299/jamdsm.2021jamdsm0069]

44. Wang, K.; Pesch, E.; Kress, D.; Fridman, I.; Boysen, N. The piggyback transportation problem: Transporting drones launched from a flying warehouse. Eur. J. Oper. Res.; 2021; 296, pp. 504-519. [DOI: https://dx.doi.org/10.1016/j.ejor.2021.03.064]

45. Pei, Z.; Dai, X.; Yuan, Y.; Du, R.; Liu, C. Managing price and fleet size for courier service with shared drones. Omega-Int. J. Manage. Sci.; 2021; 104, 102482. [DOI: https://dx.doi.org/10.1016/j.omega.2021.102482]

46. Shen, Y.; Xu, X.; Zou, B.; Wang, H. Operating policies in multi-warehouse drone delivery systems. Int. J. Prod. Res.; 2021; 59, pp. 2140-2156. [DOI: https://dx.doi.org/10.1080/00207543.2020.1756509]

47. Coindreau, M.A.; Gallay, O.; Zufferey, N. Parcel delivery cost minimization with time window constraints using trucks and drones. Networks; 2021; 78, pp. 400-420. [DOI: https://dx.doi.org/10.1002/net.22019]

48. Sawadsitang, S.; Niyato, D.; Siew, T.P.; Wang, P.; Nutanong, S. Shipper cooperation in stochastic drone delivery: A dynamic bayesian game approach. IEEE Trans. Veh. Technol.; 2021; 70, pp. 7437-7452. [DOI: https://dx.doi.org/10.1109/TVT.2021.3090992]

49. Campbell, J.F.; Corberán, Á.; Plana, I.; Sanchis, J.M.; Segura, P. Solving the length constrained k-drones rural postman problem. Eur. J. Oper. Res.; 2021; 292, pp. 60-72. [DOI: https://dx.doi.org/10.1016/j.ejor.2020.10.035]

50. Radzki, G.; Nielsen, I.; Golińska-Dawson, P.; Bocewicz, G.; Banaszak, Z. Reactive UAV fleet’s mission planning in highly dynamic and unpredictable environments. Sustainability; 2021; 13, 5228. [DOI: https://dx.doi.org/10.3390/su13095228]

51. Li, S.; Zhang, H.; Li, Z.; Liu, H. An Air Route Network Planning Model of Logistics UAV Terminal Distribution in Urban Low Altitude Airspace. Sustainability; 2021; 13, 13079. [DOI: https://dx.doi.org/10.3390/su132313079]