Supply chain is undergoing a digital transformation driven by the emergence of advanced Information and Digital Technologies (IDT) (Núñez-Merino et al., 2020). IDT of Industry 4.0 (I4.0) refers to applying technologies intensively in industry, to achieve more effective and efficient processes. These technologies are related to the digital transition, whose scope extends beyond organizational boundaries and involves intelligent supply chain and connected customers (Ghobakhloo, 2018; Gilchrist, 2016). So, IDT of I4.0 includes the application of e-commerce tools such as Electronic Data Interchange (EDI); the adoption of Advanced Manufacturing Technologies (AMT) such as Computer Aided Design (CAD), Computer Aided Manufacturing (CAM), Computer Aided Engineering (CAE), Industrial Simulation, and Enterprise Resource Planning (ERP); and the most advanced IDT such as the Internet of Things (IoT), Augmented Reality (AR), Additive Manufacturing, Blockchain, Cloud Computing, and Big-Data analytics, among others (Ghobakhloo, 2020; Gilchrist, 2016).

IDT of I4.0 can enable logistics service providers, for example to add goods for ad hoc return to a planned delivery route and to schedule ad hoc reverse logistics stops on a pre-existing route, while ensure simultaneous environmental impact and cost minimization, and customer service maximization (Agnusdei et al., 2022). IDT of I4.0 enhance logistics by improving operational efficiency and enabling real-time data analysis, which leads to optimized supply chain management and more effective decision-making. These technologies afford a better understanding of consumers and offer options to improve their satisfaction level while simultaneously providing businesses with the capability to adapt to market changes (Lasi et al., 2014).

Last-mile logistics is identified as the final stage of a Business-to-Consumer service involving goods delivery to end customers (Lim et al., 2018). As online purchases become more popular, new challenges arise for last-mile scheduling and vehicle routing services. Inefficiencies in delivery services lead to high costs for logistics providers. In turn, customers expect more flexibility in alternative collection points or the timeliness of deliveries (e.g., same-day deliveries) (Özarık et al., 2021). In contrast to traditional urban distribution, customers with small daily delivery demands and different availability schedules are widely spatially distributed, making last-mile distribution a complex and costly issue (Lim et al., 2018). Last-mile distribution is considered the most expensive, most polluting, and least efficient part of the e-commerce supply chain, accounting for between 13% and 75% of total supply chain costs (Gevaers et al., 2009). In this context, technology plays a pivotal role in mitigating negative externalities and transforming last-mile logistics into a more sustainable and efficient system. Strategies based on IDT of I4.0 use could enhance the efficiency of the last-mile logistics delivery process (Muñoz-Villamizar et al., 2021). Further, research on emerging technology use could help improve the resilience and sustainability of last-mile logistics (Agnusdei et al., 2022).

Previous literature has explored the interrelationship between various technologies and last-mile logistics, offering both theoretical insights and empirical evidence. An analysis of previous literature reviews reveals a rich tapestry of diverse contributions. De Mello Bandeira et al. (2019) conducted a systematic literature review to identify the primary vehicle types enhancing sustainability in urban last-mile distribution, while Gee et al. (2020) reviewed existing literature on the environmental implications of e-commerce. Janjevic et al. (2019) employed a combination of literature review and case study analysis to establish an integrated conceptual framework characterizing last-mile urban e-commerce distribution strategies in mature and emerging markets. Pan et al. (Pan et al., 2021) analysed literature, proposing a new conceptual framework for smart city logistics and its relevant key perspectives. Boysen et al. (2021) examined established and novel last-mile concepts, such as drones or robots, emphasizing decision problems in configuring and operating each concept, albeit with minimal consideration of I4.0 technologies. Other authors conducted an SLR linking I4.0 technologies, offering insights into other segments of the logistics chain, like first-mile logistics (Agnusdei et al., 2022). Additionally, some works explored the impact of I4.0 technologies alongside lean management, addressing the entire supply chain (Núñez-Merino et al., 2020), or focused on the digital transformation’s impact on the entire e-commerce-linked supply chain, without specific emphasis on last-mile logistics (Mashalah et al., 2022).

As observed, literature reviews exploring the dynamic interplay between the IDT of I4.0 and last-mile logistics have exhibited a range of perspectives, yet often with a narrow focus or emphasis on distinct aspects. It is evident that numerous studies delve deeper into the impact of various IDT of I4.0 on last-mile logistics. However, these valuable insights are scattered across the literature landscape, existing as isolated islands rather than cohesive narratives interconnected with one another. Our work proposes a systematic literature review that holistically addresses the effects and consequences of IDT of I4.0 for last-mile logistics while identifying existing gaps and proposing lines of future research to be undertaken.

The remainder of this work is organized as follows. The next section describes the methodology used. Then, the main results of our analysis are discussed and gaps and lines of future research are identified. Finally, some conclusions are drawn in the last section.

Systematic literature review (SLR) was used to identify, select, and evaluate published research. This method ensures a structured, repeatable, and scientific process that allows existing information to be rigorously synthesized (Denyer & Tranfield, 2009; Tranfield et al., 2003). The SLR process is composed of the following five phases (Denyer & Tranfield, 2009): (1) formulating the research question; (2) locating studies; (3) selecting and evaluating relevant studies; (4) analysing and synthesising the results obtained; and (5) reporting the results.

Starting with the first phase, this SLR aims to answer the following research question (see Figure 1):

This general question is subdivided into the following specific questions:

The second phase is the location of the literature. This step is based on the construction of search strings that allow the identification of precisely linked studies, and the selection of search engines where it is possible to explore large volumes of documents related to the subject under investigation (Denyer & Tranfield, 2009; Tranfield et al., 2003). This study used the Web of Science (WoS) search engine, a widely utilized database in operations management studies (Garcia-Buendia et al., 2021), to locate the literature on articles linking IDT of I4.0 and last-mile logistics (Núñez-Merino et al., 2020; Oliveira-Dias, Maqueira-Marín, et al., 2022).

The search strings were constructed by combining representative keywords from each of the two sets of research objects (IDT of I4.0 and last-mile logistics), taking into account synonyms and related words, and finding the intersection zone of the two sets (Table 1). The search strings allow for results that are specific enough to locate studies related to the search topic, but broad enough not to restrict possible results. Keywords were determined and selected in conjunction with the participating researchers in this study. Once the keywords were established, search strings were constructed using logical and boolean operators to ensure that the documents found contained at least one of the keywords belonging to each of the two research areas, either in the title, abstract, author keywords or expanded keywords. In addition, repeated pilot searches were carried out to refine the selection of words used, eliminating those that were not relevant or that generated ambiguous results. So, search strings were meticulously designed to retrieve documents encompassing IDT of I4.0 and last-mile logistics (see Figure 1 and Table 1), retrieving 1,254 identified articles.

In the third phase of the study, the objective was to refine the selection of relevant documents for the research. Inclusion and exclusion criteria were set to ensure the articles’ relevance. In this work, the search was initially limited to all documents obtained in the WoS database from the search strings detailed in the previous phase (inclusion criteria), and subsequently the search was limited to specific research areas: engineering, computer science, business economics, transport, operations research, management science, telecommunications, environmental science, ecology and science and technology. This first filtering reduced the results from 1,254 to 1,244 documents.

Subsequently, a language and document type filter were implemented, searching only for articles written in English (1,237 articles) and documents of type “Article”, “Early access” and “Review Article”, reducing the results from 1,237 to 828 articles. Finally, the search concentrated on papers published between 2001 and 2022 inclusive, resulting in a total of 704 articles. The search was concentrated on this period of time as the first articles related to the topic under investigation were detected from 2001 onward (George & Morin, 2001; Reich & Rosenbaum, 2001).

To ensure the quality of the selected papers, a selection was made by classifying the journals in which the articles were published according to their Journal Citation Report (JCR) Journal Impact Factor (JIF). This classification showed that, out of the 704 articles obtained in the previous step, 147 were not published in journals with impact factor, so they lacked the necessary quality. After the application of this filter, 557 articles remained.

Given the large number of articles for an SLR, we applied a more restrictive filter, following the approach of Danese, Manfè, & Romano (2018). This filter focused on the quality and reputation of the journals where the articles were published, using the Academic Journal Guide ranking system by ABS, which classifies journals from 1 to 4*, with 4* representing the highest-ranked titles. In this sense, the ABS ranking provides a more comprehensive assessment of journals compared to the JCR, as it not only considers the journal’s impact factor but also incorporates expert evaluations of its reputation. In this way, we focus only on the highest-ranked journals in this classification (levels 3, 4 and 4*) in its 2021 edition (Chartered Association of Business Schools, 2021). This left 109 articles.

Finally, the remaining articles were evaluated to determine whether the papers were closely related to the topic, discarding those that were not related to the research question (exclusion criteria). To do this, the title, abstract and keywords of each of these articles were examined, discarding a total of 12, resulting in a total of 97 articles to be analysed.

The fourth phase consisted of analyzing and synthesizing these 97 articles through a thorough reading. In this phase, a predefined codification and structured classification of the content of the articles was carried out in order to extract information of interest in each of the studies. The codification and classification parameters were extensively discussed among all researchers, reaching a consensus on which variables should be considered for coding the articles.

During the classification process, a database was implemented using an Excel spreadsheet, in which the articles’ authors, title, year, journal/source, abstract, technology used, contribution, logistics function involved, supply chain members affected and last-mile strategy used were captured for each of the 97 articles.

To group the articles, we applied a framework from existing literature, specifically focusing on the Technology Life Cycle (TLC) and I4.0 technologies (Núñez-Merino et al., 2020; Oliveira-Dias, Maqueira-Marín, et al., 2022). The TLC illustrates the progression of technology over time, marking phases of introduction, growth, maturity, and eventual decline. Technologies are considered mature when their usage becomes widespread (maturity stage) and are classified as emerging when they are newly introduced to the market (introduction and growth stages) (Oliveira-Dias, Maqueira Marín, et al., 2022; Taylor & Taylor, 2012).

The articles were grouped into research sublines according to the specific technology or groups of technologies used to further analyse the role and implications of IDT of I4.0 in last mile logistics. Gartner’s Hype Cycle for emerging technologies and previous literature on I4.0 technologies (Núñez-Merino et al., 2020; Oliveira-Dias, Maqueira-Marín, et al., 2022) were used to determine where each of the technologies is positioned in the TLC. Previous literature has found that mature technologies are enablers of efficiency and agility-oriented supply chain strategies while emerging technologies require a consolidation process to be a useful mechanism to achieve efficiency and/or agility in the supply chain (Oliveira-Dias, Maqueira Marín, et al., 2022). In this context, it would be valuable to analyse the differing roles of mature and emerging technologies in last-mile logistics.

The literature analysis and synthesis not only allowed for the identification of potential research avenues concerning the relationship between IDT of I4.0 and last-mile logistics but also highlighted existing research gaps within the various lines of research.

The final phase involved preparing a report on the results of the analysis of the selected literature to provide a general overview of the articles and their characteristics and highlight clear trends. Grouping based on the previous phase’s detailed classification identified underexplored and unexplored areas to outline how future research can address these gaps. The process was supported by the researchers discussing the findings.

Thus, our aim is to offer a comprehensive overview of what has been explored within each identified line of research, highlighting how the main contributions of individual works relate to one another. This approach allows for a clearer understanding of the added value brought by the classification and offers insight into the role of the technologies analysed in the context of last-mile logistics within each research line.

Figure 2 was prepared to illustrate the methodological procedure followed in this study and described in this section.

This section presents a descriptive analysis of the 97 identified papers on the relationship between IDT of I4.0 and last-mile logistics. It also offers a novel classification of the literature on the IDT of I4.0 and last-mile logistics, based on the technology lifecycle (TLC). This approach aligns with recent studies examining the impact of I4.0 technologies on supply chain strategies (Núñez-Merino et al., 2020; Oliveira-Dias, Maqueira-Marín, et al., 2022).

This study offers a comprehensive insight into the role of the IDT of 4.0 on last-mile logistics and presents an innovative classification of the current literature on this subject through content analysis. This classification spotlights different technologies’ utility for improving the processes involved in last-mile logistics according to the stage reached in the technology lifecycle. Moreover, the study offers a summary of existing literature regarding this subject, which researchers can utilize to promptly find pertinent material and identify potential avenues for future research.

The classification has also identified some gaps that research lines must address in the future. Mature IDT can be observed to have a limited potential for impacting last-mile logistics. As technologies such as GPS and RFID exist that complement technology roll-out by facilitating information exchange, it would be interesting to explore further the costs involved in implementing said technologies, and the opportunity to replace them or fully integrate them with other more recent technologies such as RAIN-RFID, which combines RFID technology with cloud computing. A comparison of SaaS (Software as a Service) systems and ERP needs to be established in terms of last-mile logistics efficiency. Also, a study is needed on the integration of web technologies and IoT, robots, and Blockchain along the entire logistics process to achieve higher flexibility levels in last-mile logistics contexts.

Regarding the research line on the impact of emerging IDT on last-mile logistics, no studies have been observed on the difficulties of implementing these technologies or their impact on employees in this particular context. Further exploration is therefore required of the difficulty of implementing this type of technologies and its long-term impact considering costs, quality, service level, and flexibility. In this sense, and considering recent findings (Núñez-Merino et al., 2024), it would be necessary to analyse the impact of quantum computing to optimize transportation scheduling for parcel delivery and identify the optimal routes in terms of time and distance while considering constraints such as arrival times, vehicle loads and carrier uptime. The role of employee acceptance and adaptation in the implementation of these technologies in last-mile logistics context should also be studied. It would also be interesting to analyse the impact of emerging technology sets on last-mile logistics. This would enable us to understand how one technology enhances another and how this ultimately impacts last-mile logistics processes.

Diverse future research lines are proposed to explore the impact of specific technologies on last-mile logistics. In the area of simulation, analyzing the system’s dynamic interactions and predicting its evolution are suggested, and evaluating profitability and flexibility considering common variables such as delivery times, costs, emissions, traffic, etc. Applying autonomous vehicles to picking and warehousing processes is proposed to reduce times and product localization errors. Secure delivery is a great concern with autonomous vehicles and should be studied with the integration of different technologies to enhance the security of these means of transport, such as sensors or Blockchain. AI is another technology that can be leveraged with other IDT, such as Big Data, and algorithms should be developed that consider external real-time variables, such as traffic or weather conditions, to decide on optimal delivery routes.

Regarding autonomous aerial vehicles, it is suggested that the difficulties associated with the transition to a fleet of delivery drones should be investigated considering a range of contextual factors such as cargo weight, weather conditions, population density, economic aspects, and also operative training to identify the technology’s drivers and inhibitors in the last-mile logistics. Exploring IoT applications in last-mile logistics is also proposed, particularly regarding real-time interconnection between transportation means, warehouses, and systems. In other regards, supercharging AI with Big Data is also recommended to improve real-time decision-making, considering, for example, the state of traffic or the weather to determine optimal delivery routes. Lastly, further exploration is suggested of the application of digital twins and Blockchain to last-mile logistics, especially in warehouses, by integrating other technologies such as IoT with the former and data management between logistics chain agents in the case of Blockchain.

Finally, from the general perspective of IDT’s impact on last-mile logistics, it is necessary to continue exploring the integration of information system functions with emerging technologies to achieve greater information management efficiency. Systems make structured information available and easy to process, which makes it ideal for feeding into AI systems or simulators.

Our study yields practical implications worth considering. From a managerial point-of-view, this study provides managers with a broad understanding of the IDT of I4.0 applications and how they influence on last-mile logistics. It compiles a diverse array of mature and emerging I4.0 technologies that have demonstrated their influence on various aspects of last-mile logistics. Thus, this research will assist managers in pinpointing the specific IDT of I4.0 that can enhance the processes linked to last-mile logistics. By highlighting key insights in the field, this work encourages managers and practitioners to consider implementing these technologies to improve the efficiency and flexibility of last-mile logistics operations.

Our SLR exhibits strengths as well as certain limitations. As regards the strengths, we adopted a clear and rigorous literature review approach and, in particular, a careful selection of journals following the criteria previously used in the literature (Danese et al., 2018). Regarding limitations, it is important to acknowledge that the criteria used for selecting journals may have excluded high-quality and impactful articles published in journals deemed lower in quality and reputation according to the ABS ranking. Another limitation could arise from the subjective nature of paper categorization. Nonetheless, we have followed the established phases in the SLR methodology and utilized classification taxonomies from previous studies to address this concern. Moreover, article classification was a collaborative effort involving a team of researchers, further enhancing its reliability.

The authors are grateful for the financial support provided by the University of Jaén (Action 2 of the Operational Plan for Research Support 2021-22) and the Spanish Ministry of Science, Innovation and Universities (Research Project PID2023-148345NB-I00).

All authors contributed to the conceptualization, methodology, data collection, data analysis, drafting of the original manuscript, reviewing and editing, as well as supervision.