A systematic literature review and bibliometric analysis of last-mile E-commerce delivery in urban areas

ABSTRACT

This research examines the impact of e-commerce on urban last-mile distribution through a comprehensive analysis of scientific studies. A corpus of 317 publications spanning two decades was reviewed, identifying 111 pertinent sources. Utilizing bibliometric analysis and systematic assessment, the study reveals the effects of e-commerce on last-mile delivery comprehensively. Key findings encompass environmental, economic, social, and technological impacts. Notably, while existing research has explored these aspects, the spatial impact of e-commerce remains inadequately addressed. This paper bridges this gap, contributing to the discussion on freight sustainability by elucidating current trends and outlining future research directions in last-mile e-commerce delivery. By providing insights into prevailing research trends and outlining potential avenues for further exploration, this article enriches scholarly discourse on freight sustainability.

KEYWORDS:

• E-commerce
• urban logistics
• last mile delivery
• bibliometric analysis
• literature review

1. Introduction

In the landscape of e-commerce, last-mile delivery stands out as a crucial aspect of urban freight transport, encompassing the final leg of the delivery chain from transit point to destination (Nogueira et al., 2021). With the surge in online orders, consumer expectations for swift deliveries have heightened, accentuating concerns over the environmental consequences of e-shopping preferences (Guo et al., 2019; Manerba et al., 2018). The pursuit of sustainable urban transportation grapples with nuanced challenges, balancing product consolidation against the demand for rapid deliveries (Cortes & Suzuki, 2020; Jaller & Pahwa, 2020). This escalating pressure underscores the urgent need to address environmental impacts, particularly greenhouse gas emissions from fossil fuel-powered vehicles in e-commerce logistics operations (Calvet et al., 2021; Florio et al., 2018). The complexity of the e-commerce industry is further compounded by the management of increasing order volumes with tight delivery windows spanning various distances (Gharehgozli et al., 2017). Notably, the rise of B2C e-commerce has significantly complicated the last-mile delivery process (Morganti et al., 2014). Last-mile delivery emerges as a costly, inefficient, and environmentally burdensome phase of urban logistics supply chains (Brown & Guiffrida, 2014; Lim et al., 2018; Pan et al., 2017), often constituting up to 28% of total delivery costs (Bergmann et al., 2020; Deng et al., 2021).

As e-commerce continues to proliferate, last-mile delivery becomes a growing challenge in cities worldwide, exacerbating traffic congestion, pollution, and associated health issues. Despite being the shortest segment of global freight distribution, the last mile is commonly perceived as the most daunting and expensive aspect of e-commerce supply chain management (Ren et al., 2020). City-centric last-mile delivery management faces formidable hurdles, including increased fuel consumption, delivery delays, and reduced efficiency due to traffic congestion (Ranieri et al., 2018). While the potential impacts of expanding e-commerce on transportation, mobility, and land use are acknowledged, their integration into planning practice remains limited. This study endeavors to deepen the understanding of how last-mile e-commerce distribution influences urban areas.

The objective is achieved by addressing the subsequent research questions: 1. What are the key areas of research in the field of urban freight last-mile e-commerce delivery? 2. What are the associated impacts in terms of traffic, environment, economy, technology and spatial distribution of e-commerce last mile delivery in urban areas. The next section (section 3) addresses the first objective, while section 4 comprehensively explores the multifaceted dimensions of the second objective.

2. Systematic literature review methodology

The paper follows the PRISMA methodology on last-mile delivery for e-commerce and its impacts on conducting a systematic literature review. In accordance with the PRISMA guideline, the review is divided into four stages: identification, screening, eligibility, and inclusion. To ensure a comprehensive and unbiased search, a refined search strategy was employed across three academic databases: Scopus, Web of Science, and Google Scholar. The search terms incorporated Boolean operators (AND, OR, NOT) to combine keywords strategically and optimize search results. The following refined search string was utilized: (TITLE-ABS-KEY (‘last mile delivery’ OR ‘last mile distribution’) AND (e-commerce OR online shopping) AND (urban OR city OR metropolitan)) NOT (reviews OR conference abstracts OR editorial materials). This search string targets studies with ‘last-mile delivery’ or ‘last-mile distribution’ in the title, abstract, or keywords, along with ‘e-commerce’ or ‘online shopping’ to ensure relevance to the research topic. The initial search identified 317 documents. To focus on scholarly articles relevant to the urban context of last-mile e-commerce delivery, the search terms included ‘urban’, ‘city’, or ‘metropolitan’ and excluded all other works such as published/unpublished reports, master’s theses, and doctoral dissertations. Reviews, conference abstracts, and editorial materials were also excluded to refine the results and prioritize scholarly articles. After applying these filters, a rigorous quality assessment using pre-defined inclusion criteria yielded a final selection of 111 documents directly relevant to the research topic, as shown in Figure 1, and detailed in Table 1.

Figure 1. Publications selection methodology.

Table 1. Inclusion criteria for paper selection.

Database	Scopus, Web of Science, Google Scholar
Analyzed Field	Title, Abstract, Keywords
Search Phrase	TITLE-ABS-KEY (“impacts AND “e-commerce” AND “last mile delivery” OR “last mile distribution”
Document Types	Articles from Journals
Language	English
Source Type	Peer Reviewed
Time Interval	2000–2022
Final number of articles analyzed for the review	111

Further, a bibliometric analysis of documents shows clustering of research by country/region of research in the documents to figure out the regions where the studies are concentrated. The review was also carried out with respect to the most-used keywords in the collected documents, which throw light on the prominent themes and focal points of the research landscape. This was processed with VOS viewer software, which analyze how often these ideas appear together. The software puts the ideas into groups called ‘clusters’. In the co-occurrence graph generated by software, each cluster has a central node that brings together many ideas. These ideas are shown as large spheres.

3. Literature analysis

The section provides a comprehensive approach to literature analysis, incorporating two key components: document analysis and bibliometric analysis. These two components are further subdivided.

3.1. Analysis of documents

The selected articles were classified based on annual publication trends, subject areas, research methodologies, journal and publisher allocations.

3.1.1. Chronological publishing trend

The number of papers published each year in the portfolio from 2002 to 2022 is shown in Figure 2. A significant increase in the number of publications on e-commerce last mile logistics research can be observed from year 2012. A total of 111 articles were searchable by 2021. The active growth of academic research suggests that last-mile e-commerce delivery is expanding in both scope and branch. Additionally, it is clear from the publications and recent discussions of e-commerce last-mile delivery that social demand, market acceptance, public awareness, and actual use of sustainable logistics practices are all rising significantly.

Figure 2. Year-wise publication.

3.1.2. Classification by subject area

The documents were put into five research categories based on what they were about. The major focus of the literature is on transportation and logistics (43,39%), followed by social science and economy (25,23%), supply chain management (21, 19%), sustainability (15, 14%), and lastly technology and engineering (7, 6%). The most prevalent themes in the literature review are depicted in Figure 3.

Figure 3. Publications by subject area.

3.1.3. Journal allocation analysis

A total of 111 articles were identified across 81 different journals. Among these journals, eight journals, each with a minimum of three articles, collectively contribute to 36% of the total publications. The top-ranking journals in terms of the number of publications are: Journal of Cleaner Production (8 articles, 7.2%), Transport Research Procedia (6 articles, 5.4%), International Journal of Physical Distribution & Logistics Management (6 articles, 5.4%), Transportation Research Part D (5 articles, 4.5%), Sustainability (5 articles, 4.2%), Research in Transportation Business and Management (4 articles, 3.6%), and International Journal of Production Economics (3 articles, 2.7%). Figure 4 represents the distribution of publications by journal names.

Figure 4. Number of publications by journal name.

3.1.4. Main research methodologies

The broad research methodologies identified through the literature were: ‘review’, ‘qualitative and conceptual paper’, ‘modeling and simulation’, ‘questionnaire survey’, ‘empirical paper’, and ‘hybrid techniques’. The Endnote smart group function was used to automatically categorize the indexed records based on their title, abstract, and Keywords. Modeling and simulation is the most utilized technique, followed by hybrid research methods as shown in Table 2.

Table 2. Description of research methodologies.

Research Methodology	Description	Count
Review	A study that reviews and analyses the progress of current research	20
Qualitative and conceptual	A study that reviews and analyses the progress of current research	13
Modelling and simulation	A study that uses a model for mathematical functions for the decision-making process	42
Questionnaire survey	A research instrument consisting of a series of questions (or other types of prompts) for the purpose of gathering information from respondents	8
Empirical research	A study of gaining solution proposal by means of direct and indirect case study or experience	7
Hybrid techniques	A study that uses a mixture two or more methods describe above	21

3.1.5. Classification by publisher

Publications were also categorized based on their publishers. Notably, Elsevier holds the highest share of publications, followed by Emerald, and Taylor and Francis as shown in Figure 5.

Figure 5. Number of publications as per publisher.

3.2. Bibliometric analysis

3.2.1. Analysis of author’s country

Many studies have been done on the last mile of distribution for e-commerce, especially in Europe, Asia, and North America, as shown in Figure 6. Italy, Germany, China, the United Kingdom, and the United States have the most publications, but the United States has the most total citations. Other countries or regions, like France, Brazil, Belgium, and India, have fewer publications, but their average normalized citation numbers are still high, which shows how influential they are. Most of the studies from these countries or regions were also published in the last three years, which shows that they are getting more involved in promoting research.

Figure 6. Mapping of countries contribute to e-commerce last mile delivery research.

3.2.2. Analysis of keyword Co-occurrence

The analysis of keyword co-occurrence employs a content analysis technique that utilizes author-assigned keywords to construct semantic visual maps, unveiling the cognitive structure within the research domain. To achieve this, the bibliometric map was generated using the VOSviewer software, leveraging co-occurrence data and the associated strength of relationships. The resulting visual representation employs nodes and links to depict keyword relationships. In this context, nodes symbolize individual keywords, with their size proportional to the frequency of appearance in the documents. Larger nodes indicate higher frequency, while smaller nodes correspond to lower frequency. The links, or edges, illustrate relationships between pairs of nodes. The width of these links reflects the strength of the relationship, with greater width signifying a stronger association.

An analysis of the co-occurrence of keywords was performed to define the internal organization and composition of the literature on e-commerce. To acquire a comprehensive intellectual landscape of the topic of e-commerce last-mile delivery, the options ‘All Keywords’ and ‘Full Counting’ in VOSviewer analysis were considered. A network of 60 nodes representing keywords (1107 keywords in all documents) and 822 linkages was created by setting the minimum number of occurrences of each term to 5, as shown in Figure 7.

Figure 7. Co-occurrence analysis of keywords.

3.2.3. Keyword-cluster analysis

The keyword co-occurrence network map has been segmented into four distinct clusters, each denoted by a unique color. Cluster 1 emphasizes various themes in e-commerce last-mile logistics, with a predominant focus on distribution network architecture and technological advancements. Cluster 2 predominantly addresses the sustainability of e-commerce, encompassing studies on the environmental, economic, and social implications of online shopping expansion. Cluster 3 predominantly examines the environmental impact of last-mile e-commerce delivery, with a focus on carbon emissions and sustainable delivery methods. Lastly, Cluster 4 delves into e-commerce last-mile delivery strategies, exploring approaches such as electric cargo bikes, parcel lockers, and sustainable city logistics.

3.2.3.1. Cluster-1: E-commerce logistics

The first cluster focuses on various themes in e-commerce last-mile logistics. Academic research on distribution network architecture from a technological perspective dominated. Global e-commerce last mile logistics research networks, trends, and advanced theories were explored in (Hu et al., 2019). (Cardenas et al., 2017; Milewski, 2019) created an external cost delivery index (Visser & Lanzendorf, 2004) examines how B2C e-commerce affects mobility and accessibility. (Rotem-Mindali & Weltevreden, 2013) extensively reviews the impact of business-to-consumer (B2C) e-commerce on mobility, focusing on both personal travel and freight transport. Time-windowed vehicle routing and scheduling models, traffic simulation, and other methodologies show how e-commerce affects urban freight transportation and the environment (Breunig et al., 2019; Taniguchi & Kakimoto, 2004). Other studies reviewed e-commerce city logistics literature on diverse topics (Delfmann et al., 2002; Dolati Neghabadi et al., 2019). Big data solutions assist city logistics plan design and assessment (Taniguchi et al., 2016). Online and conventional shopping were contrasted (Farag et al., 2006; Weltevreden & Rietbergen, 2007, 2009). E-commerce research suggest that logistics knowledge boosts firm performance (Aziz et al., 2020).

3.2.3.2. Cluster-2: sustainability of E-commerce

Cluster 2 includes mostly review-based studies. However, the sustainability of e-commerce is the main area of research in this cluster. Online shopping expansion and its implications on last-mile deliveries, including traffic, the economy, and the environment, were examined (Chaabane et al., 2012) emphasized sustainable supplier chains (Palsson et al., 2017) examined the energy use of storefronts and home delivery models in a literature study. Empirical and review research examines sustainable last-mile delivery (Nogueira et al., 2021) (Bjørgen et al., 2021) explored the impact of home delivery services on travel and city planning, as well as the link between online shopping and travel habits (Nürnberg, 2019; Schliwa et al., 2015) examines cargo cycles’ potential to improve city logistics sustainability and encourage their use. E-commerce’s regional implications on travel and fuel utilization were examined by (Stinson et al., 2019) (Buldeo Rai et al., 2019) examine customers’ willingness to adopt sustainable last-mile solutions and how to build them to remain desirable. Studies have examined the problems and applications of electric cars in last-mile e-commerce delivery (Anosike et al., 2021; Duarte et al., 2016).

3.2.3.3. Cluster-3: environmental impact of last-mile e-commerce delivery

Studies in the third cluster mostly focus on the environmental impacts of e-commerce (Van Duin et al., 2020). examined sustainable e-commerce delivery methods and the environmental impacts of online shopping. Logistically (Demir et al., 2015; Mangiaracina et al., 2015), evaluate B2C e-commerce environmental sustainability literature (Brown & Guiffrida, 2014; Shahmohammadi et al., 2020) examined the carbon emissions from last-mile delivery to client pickup in traditional shopping. Another research (van Loon et al., 2014) contrasts online and in-store shopping’s carbon footprint (Van Loon et al., 2015) analyses the pros and cons of several e-retail channels and provides a methodology for further investigation into the environmental viability of online retailing of FMCG products. A study (Edwards et al., 2010) explored how much CO2 is produced by last-mile e-commerce deliveries and personal shopping trips. It also compares carbon audits of traditional and online retail supply chains and models to analyze the environmental impacts of several e-fulfillment options.

3.2.3.4. Cluster-4: E-commerce last mile delivery strategies

Most research in the fourth cluster focuses on last-mile e-commerce delivery approaches such as electric cargo bikes, parcel lockers, light goods vehicles, and sustainable city logistics. In an urban distribution system, combining first- and last-mile delivery operations affects route efficiency (Bergmann et al., 2020). An integrated methodology compares urban last-mile e-commerce delivery techniques for established and emerging economies (Janjevic & Winkenbach, 2020). Price structures of numerous Antwerp urban e-commerce distribution scenarios were examined (Arnold et al., 2018) (Zhang et al., 2019) describes how a two-echelon logistics distribution strategy for online merchants may handle last-mile split delivery (Yuen et al., 2018) examines client willingness to employ self-collection for last-mile delivery (Ranieri et al., 2018) reviewed the latest scientific literature on new last-mile logistics methods to reduce externality costs. Parcel lockers may be better than home delivery (Van Duin et al., 2020). To boost e-grocery home delivery (Pan et al., 2017) proposes a customer-data-driven strategy.

4. Literature findings

Based on the literature corpus reviewed for this paper, the impact of e-commerce-related last mile logistics systems can be classified as traffic, economic, environmental, technological and spatial impacts. This section discusses the key findings emerging from literature, under these five categories.

4.1. Traffic impacts

The choice to delve into traffic impacts stems from the significant role of e-commerce last-mile delivery in shaping urban transportation dynamics. By analyzing traffic impacts, we aim to understand the implications of increased delivery activities on vehicle kilometers traveled (VKT), traffic congestion, and environmental concerns such as emissions. Traffic impacts are one of the most common impacts of e-commerce-related last-mile deliveries Increased e-commerce could directly lead to more vehicle kilometers travelled (VKT) due to smaller fleet size and longer delivery being used for home deliveries (van Loon et al., 2014). The widely held belief that shopping online is more environmentally friendly than going to a store is based on the amount of VKT for each purchase. Empty return trips to warehouses are common (Edwards et al., 2010). Residential delivery mostly relies on fuel-intensive vans, emitting more per tonne than larger vehicles (Allen et al., 2012). Quick home delivery is also required to make e-commerce more appealing to consumers, resulting in inefficient routes and low loading factor (Pettersson et al., 2018; Sui & Rejeski, 2002; Visser et al., 2014). Alternatively, e-commerce delivery vehicles might replace retail store shopping trips. However, there is little scientific information on how the effects would influence consumer mobility. It is also unclear whether e-commerce necessarily results in reduced passenger movement, and if it does, home delivery through a motorized mode may replace a trip that would have been made by a sustainable mode (Visser & Lanzendorf, 2004). Trip chaining and optimizing last-mile delivery routes can mitigate traffic and environmental impacts (Brown & Guiffrida, 2014). Increasing deliveries per trip and shifting shopping habits might reduce vehicle usage. For instance, several smaller deliveries per week might replace a weekly grocery trip (Jaller & Pahwa, 2020) highlighted that the E-commerce efficiency depends on delivery round size; over 92 deliveries make e-commerce favorable. E-shopping may increase shopping frequency, resulting in more frequent but smaller deliveries replacing traditional weekly grocery trips. The impact of increased home deliveries on traffic patterns is a societal concern.

Studies (Cao et al., 2010; Le et al., 2022; Visser & Lanzendorf, 2004; Weltevreden & Rotem-Mindali, 2009) disagree on the effects, with some suggesting that e-commerce delivery could raise freight traffic while reducing shopping-related passenger traffic, while other studies present opposing views. Rural deliveries can be three times more expensive than urban ones (Boyer et al., 2009; Gevaers et al., 2014). While metropolitan areas offer logistic advantages due to higher density and lower costs, while also experiencing greater adverse effects such as congestion, noise, and pollutants (Holguín-Veras et al., 2006)., Extending delivery windows reduces emissions per order and enhances efficiency. In London (Edwards et al., 2009), found that elongating residential delivery windows from 3 hours to 3 hours 45 minutes, and 6 hours lowered transport expenses by 6–12% and 17–24% respectively. These savings likely connect to reduced VKT and carbon emissions (Mangiaracina et al., 2015). highlighted that e-commerce last-mile delivery is considered less environment friendly and faces several challenges. Initial daytime distribution clashes with most consumers’ work schedules, leads to failed deliveries (Visser & Lanzendorf, 2004; Visser et al., 2014). Failed deliveries result in repeated attempts, with rates increasing for higher residential delivery volumes (Weltevreden & Rotem-Mindali, 2009; Weltevreden & Van Rietbergen, 2009). Next-day delivery adds complexity, hindering route efficiency and increasing vehicle use (Allen et al., 2017, 2018). Delivery vans are often underloaded in capacity and weight (Allen et al., 2017, 2021), against recommended rates of under 30% in urban and 50% in nonurban areas (Edwards et al., 2009; Meschi et al., 2011). Moreover, vehicles often return to the warehouse empty (Edwards et al., 2010). Residential delivery primarily employs light cargo vehicles or vans, which use more fuel and emit more emissions per metric tonne than larger vehicles (Allen et al., 2012, 2017). Utilizing sustainable modes for last-mile delivery is crucial to cut the carbon footprint. E-commerce logistic companies can minimize the VKT and subsequently the environmental footprint of delivery by increasing the number of deliveries per delivery round and using alternative sustainable mode of delivery (Janjevic & Winkenbach, 2020; Zacharias & Zhang, 2015).

4.2. Environmental impacts

Environmental considerations are paramount in assessing the sustainability of e-commerce last-mile delivery. This section explores the environmental footprint of delivery operations, including CO2 emissions, energy consumption, and the ecological consequences of packaging and returns. Understanding these impacts is crucial for devising strategies to mitigate environmental harm. The environmental impact of e-commerce last-mile delivery varies based on factors like item quantity, drop density, pickup windows, returns, and failed deliveries (van Loon et al., 2014). E-commerce’s growth has led to more store visits for product testing, compounded by generous return policies that incentivize online orders (Buldeo Rai et al., 2019; Pei et al., 2014). Regulations enabling faster online sales growth than traditional stores have fuelled this trend (Hua et al., 2017). Returns, requiring energy for shipping, re-manufacturing, and packaging, adversely affect the environment (Hua et al., 2017). The surge in B2C e-commerce raises the demand for home delivery, increasing social and environmental transportation costs (Taniguchi et al., 2016). Addressing environmental issues linked to urban commodity distribution has gained traction in the last two decades (Morganti et al., 2014; Yuen et al., 2018). The shift in urban freight patterns due to online shopping may not necessarily worsen environmental impact compared to traditional shopping, especially with varied delivery methods like home delivery, pickup locations, or click and collect systems (Cravioto et al., 2013). Environmental concerns like traffic, CO2 emissions, and congestion have risen alongside e-commerce’s growth. Last-mile delivery demand is predicted to increase by 78% globally by 2030 (Mucowska, 2021), driven by urban delivery developments and services such as e-grocers and food delivery. Urban freight transport plays a pivotal role, particularly in road transport, causing congestion, noise, and air pollution (Ranieri et al., 2018). Despite consumer preference for home delivery, increased awareness of environmental costs is emerging (Morganti et al., 2014; Yuen et al., 2018) and urban transportation infrastructure remains largely unchanged (Marujo et al., 2018). Fast delivery models may lead to increased greenhouse gas emissions, as quicker delivery options encourage more vehicle use (Ding & Jin, 2019). A study (Lin et al., 2018) explored the environmental and cost implications of on-demand same-day delivery (SDD) services in e-commerce. It compares three delivery paradigms – hub-and-spoke, SDD with a commercial fleet, and SDD by crowdsourcing. The study finds that crowdsourcing is a promising, cost-effective solution for same-day delivery, but it may lead to increased fuel consumption and emissions during peak demand due to additional vehicle detours.

It’s important to mitigate this impact, potentially through differential pricing or incentives (Guo et al., 2019; Manerba et al., 2018). Research emphasizes the importance of reducing GHG emissions from fossil fuel vehicles in supply chain logistics operations for environmental sustainability (Ehmke et al., 2012). The rise of e-commerce has led to dynamic logistics changes, especially during events like the COVID-19 pandemic (Bjørgen et al., 2021). Amazon has introduced eco-friendly options like ‘no-rush shipping’ to address environmental concerns (Balan & Zulnasree, 2021). The environmental sustainability of last-mile delivery primarily hinges on road freight transport’s impact, including pollution, greenhouse gases, noise, and congestion (Demir et al., 2015). The CO2 emissions associated with e-commerce’s last mile are notably high (Seebauer et al., 2016), necessitating reduced emissions in logistical operations (Savelsbergh & van Woensel, 2016). Using cargo bikes instead of vans for the last mile can significantly decrease greenhouse gas emissions (Wang et al., 2019). Optimized route planning is a common approach to reduce emissions, vehicle use, energy consumption, and CO2 emissions in urban freight distribution (Ehmke et al., 2012; Köster et al., 2018; Letnik et al., 2018; Reyes et al., 2017). A combination of measures, including cargo bikes and clean fuels, can significantly decrease energy use and carbon footprints (Arnold et al., 2018; Schliwa et al., 2015). The overall environmental impact of online shopping is influenced by factors like purchase frequency, transportation mode, and changes in consumer behavior (Letnik et al., 2018).

4.3. Economic impacts

Examining the economic dimensions of e-commerce last-mile delivery provides insights into cost structures, profitability, and market dynamics. By scrutinizing factors such as operational costs, delivery options, and consumer behavior, we aim to elucidate the financial implications for businesses, consumers, and society at large. The focus on economic sustainability in e-commerce largely revolves around cost reduction (Dutta et al., 2020; Nikolaou et al., 2013). Studies have examined operational, transport, and shipping costs, along with external costs that encompass factors like traffic congestion, accidents, air pollution, noise, and climate change, often more pronounced in cities (Cardenas et al., 2017; Vanelslander et al., 2013). External costs contribute approximately 23% to operational costs in home deliveries using vans (Arnold et al., 2018). Shipping costs can also be mitigated through parcel package redesigns, which can have environmental benefits (Sarkis et al., 2004). Despite lower inventory control costs, e-commerce faces higher return costs (Milewski, 2019). Delivery speed, options, and costs significantly influence the profitability of last-mile delivery (Nguyen et al., 2019). Customers prioritize delivery costs, followed by delivery speed due to rising expectations for faster delivery times (Clausen et al., 2016; Collier & Bienstock, 2006; Garver et al., 2012). Various delivery options, including specific days or times, further impact choice (Nguyen et al., 2019). Three last-mile delivery options are attended delivery, unattended delivery, and delivery at pick-up locations (Ignat & Chankov, 2020). Algorithms optimizing distribution centers based on intermediate sites have been proposed (Zhang et al., 2019; Zhou et al., 2018), enabling combined shipments and distribution, potentially lowering delivery costs by 5% to 16% (Zhou et al., 2018). However, the economic impact is influenced by customer density (Huang et al., 2018). Consideration of operational and financial sustainability is vital for freight distribution scenarios (Chen et al., 2017; Fancello et al., 2017; Malik et al., 2017). Authors have examined shifts from car-based to client pickup systems, which benefits logistics operators economically (Alves et al., 2019; Kedia et al., 2019; Perboli et al., 2018; Yuen et al., 2018). Online sales often offer lower costs than traditional stores, benefiting consumers and retailers. Retailers save through higher-margin products, wider geographic reach, and enhanced distribution center and stock management returns (He et al., 2019; Yao et al., 2009). However, customers might resist home delivery and online shopping costs (Goethals et al., 2012).

4.4. Technological impacts

Technological advancements play a pivotal role in shaping the efficiency and sustainability of last-mile delivery systems. This section investigates the adoption of emerging technologies, such as electric vehicles, drones, and automated logistics solutions, and their impact on operational efficiency, environmental performance, and consumer experience. It comprises communication platforms, vehicle monitoring, and route planning tools, with the goal of integrating digital platforms seamlessly for overall sustainability (Farooq et al., 2019; Nürnberg, 2019). A study (Oliveira et al., 2017) explores the potential demand for e-commerce automated delivery stations (ADS), aiming to address challenges in e-commerce home deliveries in Brazil. Electric vehicles (EVs) offer eco-friendly alternatives to diesel and gasoline vehicles, emitting no tailpipe emissions and reducing emissions by up to 30% with less noise (Anosike et al., 2021; Calvet et al., 2021). Another study (Siragusa et al., 2022) compared electric vehicles (EVs) and internal combustion engine vehicles (ICEVs) for last-mile deliveries in B2C e-commerce, revealed that despite higher initial costs, EVs are economically beneficial over an 8-year period due to lower operating expenses. Challenges include fleet size, schedules, and infrastructure adjustments (Anosike et al., 2021). Urban consolidation centers (UCC) and distribution centers (UDC) are proposed for future e-commerce freight needs, with intelligent transportation systems and underground freight concepts emerging as innovative solutions (Allen et al., 2012; Hu et al., 2019; Muñuzuri et al., 2012). Innovations such as UCCs, electric road networks, and intelligent transportation systems are supported by techniques like big data, IoT, and cloud computing (Ren et al., 2020). EV technology, widely used in passenger and freight transportation, brings high benefits to carbon emission reduction in logistics (Duarte et al., 2016). Notable technological advancements include autonomous vehicles, drones, and automated lockers (Duarte et al., 2016). Data-driven innovations improve last-mile logistics efficiency (Duarte et al., 2016). Innovations are categorized into new vehicles, proximity points, collaborative logistics, improved transport management, and routing to reduce externalities (Ranieri et al., 2018). Delivery lockers and crowd logistics represent process innovations that rely on real-time sensor data for optimal transport networks (Alnaggar et al., 2021; Kin et al., 2018). Drone-assisted deliveries and truck-drone hybrids are evolving for reliable last-mile logistics (Banyai, 2018). Drones offer the potential to transform traditional last-mile logistic delivery services (Savelsbergh & van Woensel, 2016).

4.5. Spatial impacts

The spatial dimension of e-commerce last-mile delivery elucidates the geographical patterns and urban transformations induced by online shopping trends. By examining the spatial distribution of delivery activities, retail landscapes, and consumer behavior, we aim to uncover the spatial implications for urban development, accessibility, and land use planning. While substantial research has been conducted on various impacts of e-commerce related to last-mile deliveries, the exploration of spatial impacts remains limited within the reviewed literature. The focus has primarily illuminated the potential influence of e-commerce on retail commerce and urban retail areas. For instance (Farag et al., 2006), indicates that convenient access to physical stores could deter individuals from making online purchases. Similarly (Weltevreden & Rietbergen, 2007), underscores the importance of considering diverse retail categories in different locations due to the varying e-commerce potential across product types. City districts emerge as key areas susceptible to the effects of the shift towards online shopping. A study examining the influence of e-commerce on Dutch inner cities (Boschma & Weltevreden, 2005) reveals that sectors with high e-commerce penetration, predominantly concentrate within the city zones. This concentration implies that these districts are most likely to experience significant e-commerce impacts. Customers can conveniently choose to order products online and collect them from local stores or opt for home delivery, often favouring inner city districts due to the ease of ‘trip chaining’. Moreover, an observable trend involves the decentralization of distribution systems in B2C e-commerce, resulting in the expansion of facilities to suburban and peripheral regions (Visser & Lanzendorf, 2004). In the realm of an omni-channel retail environment (Friederiszick & Glowicka, 2016), argues that traditional brick-and-mortar merchants must either downsize their physical store networks or reimagine their functions. The advent of e-commerce and remote work facilitated by the internet also introduces the potential for individuals to reside in remote areas, leading to indirect accessibility implications. This notion aligns with the ‘end of geography’ perspective, suggesting that geographical proximity to a store becomes less crucial in an era where online ordering and home delivery are prevalent (E. J. Visser & Lanzendorf, 2004). Additionally, a review study (Cao, 2009) underscores the importance of advancing research methodologies for a comprehensive understanding of the relationships among e-shopping, spatial attitudes, and travel behavior. However, separating the impact of e-commerce from other trends in retailing proves challenging (Pettersson et al., 2018). Kumar andChidambara (2023) highlight the emergence of Q-commerce in Indian cities, introducing new spatial and logistical challenges. Their study reveals that Q-commerce has led to the proliferation of ‘dark stores’—micro-fulfillment centers in residential neighborhoods for rapid delivery. These stores often operate in non-conforming areas, causing urban planning issues like increased local freight traffic, congestion, noise, and emissions, which current planning documents inadequately address. The study emphasizes the need for revised planning provisions to ensure compliance with regulations, adequate parking, and land use compatibility. Proactive spatial planning and collaboration between urban planners, local authorities, and logistics operators are essential to balance efficiency, sustainability, and urban livability as Q-commerce expands.

4.6. Research gap

The systematic literature review and bibliometric analysis presented in this study shed light on the complex landscape of last-mile e-commerce delivery in urban environments. However, several critical research gaps emerge from the analysis. Firstly, there’s a need for exploration into the integration of reverse logistics within last-mile delivery systems to optimize processes like product returns and recycling. Secondly, understanding the reciprocal relationship between emerging technologies and last-mile delivery systems is essential for future innovations. Additionally, comprehensive spatial analysis beyond the impact on brick-and-mortar businesses is necessary to assess the broader implications of urban accessibility and spatial distribution. Moreover, exploring the unique challenges and opportunities in emerging markets can provide context-specific strategies for sustainable urban development. Lastly, assessing the long-term implications of last-mile e-commerce delivery on sustainable urban development is crucial for informing urban policies and practices. Addressing these research gaps can enhance understanding of the multifaceted impacts of last-mile e-commerce delivery and contribute to the development of sustainable urban environments.

5. Discussion and conclusion

This systematic literature review and bibliometric analysis illuminate a dynamic research landscape surrounding last-mile e-commerce delivery in urban settings. A notable geographic bias is evident, with a concentration of studies in developed economies, particularly the United States. While contributions from China, Latin America, and India are present, they are fewer in number. This gap highlights the need for future research to investigate the unique challenges and opportunities faced by emerging economies, considering their distinct supply chains, logistical infrastructure, and e-commerce trends. Keyword co-occurrence analysis revealed four thematic clusters, with sustainability concerns and the implications for e-commerce recurring throughout the research. The analysis further categorized the impacts of last-mile delivery into five dimensions: traffic, environment, economy, technology, and spatial distribution. These categories, however, exhibit varying degrees of research focus. Environmental impact has received significant research attention, particularly regarding its complex relationship with traffic congestion. However, the analysis of economic impacts remains less comprehensive. Similarly, while technological advancements in delivery efficiency are well-explored, the reciprocal influence on technology itself is understudied. Spatial impacts are the least investigated area, with a primary focus on e-commerce’s effect on brick-and-mortar stores. A more comprehensive understanding is needed regarding the intricate relationship between e-commerce growth, accessibility, spatial distribution, location choices, and urban footprints. The pressure for rapid deliveries necessitates close proximity between final distribution points and customers, creating trade-offs in transportation, rental costs, and delivery speed. These factors have the potential to significantly impact urban dynamics through traffic and environmental conditions.

There are several promising avenues for future research. Firstly, investigating the integration of reverse logistics into last-mile delivery systems holds immense potential. Studies exploring strategies to optimize reverse logistics processes, such as product returns and recycling, could significantly reduce the environmental footprint and enhance urban delivery network efficiency. Secondly, a deeper understanding of the reciprocal relationship between technological advancements and last-mile delivery is crucial. Research delving into how emerging technologies like autonomous vehicles, drones, and smart delivery systems impact efficiency, sustainability, and the consumer experience can pave the way for innovative solutions. A comprehensive spatial analysis represents another critical area for future exploration. Beyond the impact on brick-and-mortar businesses, studies should investigate the broader implications of last-mile delivery on urban dynamics. This includes assessing the relationship between e-commerce expansion, urban accessibility, spatial distribution, location preferences, and area footprints. These efforts can provide valuable insights into the multifaceted impacts of last-mile delivery on transportation infrastructure, land use planning, and environmental sustainability. Furthermore, focusing on emerging market perspectives in last-mile e-commerce delivery is essential as e-commerce continues its global expansion. Understanding the unique challenges and opportunities in these markets requires research into the distinct supply chain dynamics, logistical infrastructure constraints, and socio-economic factors shaping last-mile delivery in developing regions. By developing context-specific strategies, researchers can contribute to addressing the evolving needs of urban populations in these areas. Finally, assessing the long-term implications of last-mile e-commerce delivery on sustainable urban development should be prioritized in future research. Studies that integrate last-mile delivery systems with broader urban planning initiatives can help reduce traffic congestion, minimize environmental impacts, and enhance overall quality of life in cities. By considering the interplay between transportation, land use, and economic development, scholars can contribute to creating more resilient and livable urban environments for future generations.

In conclusion, this systematic literature review and bibliometric analysis highlight the need for extensive research to comprehend the diverse impacts of last-mile e-commerce delivery on metropolitan areas, particularly in emerging countries. As urban environments evolve, a comprehensive understanding of the consequences is essential for planning sustainable urban development.

Disclosure statement

No potential conflict of interest was reported by the author(s).