Freight patterns and spatial planning requirements of third generation E-commerce in Indian cities

ABSTRACT

The increasing trend of multi-channel urban consumption, driven by the rise of E-commerce, has led to the emergence of new freight facilities. The COVID-19 pandemic further amplified this shift, with consumers relying more heavily on online purchases for regular needs, resulting in the advent of Q-commerce as a distinct sub-vertical within the E-commerce industry. However, there is a significant lack of understanding regarding the characteristics, requirements, and impacts of this sector in both existing literature and planning documents. Therefore, the aim of this research is to investigate the freight patterns and associated externalities within urban areas for this emerging trend in E-commerce, while also assessing the adequacy of planning documents in addressing these issues. The paper presents a case study-based examination of various Q-commerce grocery companies operating in Delhi, delving into their operations, freight patterns, and neighborhood-level externalities of last-mile delivery. By highlighting the externalities and deficiencies related to E-commerce, this study provides valuable insights into the planning implications of Q-commerce and its impact on urban areas.

KEYWORDS:

• E-commerce
• Q-commerce
• Hyperlocal delivery
• Urban freight
• City Logistics
• Dark Store

1. Introduction

E-commerce has experienced a remarkable surge in growth and development globally, revolutionizing the way people shop and do business. With the advent of the internet and advancements in technology, online retail has become an integral part of the modern consumer experience. From small-scale entrepreneurs to multinational corporations, businesses have capitalized on the immense opportunities offered by E-commerce to reach a wider customer base and streamline their operations. Due to this, E-commerce is transforming the retail industry and is now a major driver of global innovation and economic growth. In India, the E-commerce market has been steadily growing, and it is projected to become the world’s second-largest E-commerce market by 2034 (IBEF, 2022). The E-commerce industry in India is expected to reach $99 billion by 2024, primarily driven by the growth in the grocery and apparel sectors (Bora, 2021). The demand for online retail saw a 36% increase during the last quarter of 2020 compared to the same period in 2019. The penetration of online retail is predicted to reach 10.7% by 2024, and the number of online buyers in India is projected to reach 220 million by 2025 (IBEF, 2022). While metro cities, especially Tier I cities (population size: 1 million and more), currently handle a significant portion of E-commerce shipments, Tier-II (population size: between 100,000 to 999,999) and Tier III+ cities (population size: between 20,000 to 99,999) have emerged as major contributors, surpassing the metro cities (Unicomerce, 2021).

The E-commerce industry is constantly evolving to meet consumer demands for faster delivery. This was propelled by COVID-19 pandemic, which made a significant impact on the Indian E-commerce market, by bringing in about a surge in online shopping (Ghose et al., 2021). The surge was observed in both daily needs and non-essential items (Habib & Hamadneh, 2021; Sardjono et al., 2021). The shift was particularly noticeable in essential goods like groceries, medicines, and personal care products, with E-commerce platforms playing a crucial role in ensuring their availability during the lockdown period (Statista, 2023). Moreover, the pandemic led to an expansion of E-commerce players as traditional retailers and local businesses ventured into the online space to reach their customers in the face of physical store closures. It accelerated the shift to online shopping, leading to the emergence of a new sub-vertical in the E-commerce industry known as ‘third-generation E-commerce’. Although ‘third-generation E-commerce’ is not a widely recognized term references of it are made in the context of the Indian E-commerce section (Joseph, 2023). A new type of E-commerce known for its speedy delivery is referred to as ‘Q-commerce’ or ‘quick commerce’ (Osheen, 2022; Ranjekar & Roy, 2023; Stojanov, 2022), which is used interchangeably with other terms such as:

‘hyperlocal delivery’ (Goritiyal, 2023; Guru et al., 2023; John, 2023; Sanghi et al., 2023), and ‘on-demand delivery’ (Duthie et al., 2023; Seghezzi et al., 2021). To preserve consistency throughout this paper, the term ‘Q-commerce’ is used to represent all of the terms listed above. These businesses typically offer delivery within minutes or hours, rather than days or weeks. This makes them ideal for customers who need something quickly, such as groceries, food, or medicine. The growth of Q-commerce has had a significant impact on the retail industry, as more and more consumers become comfortable shopping online, and as the technology improves, it is expected to see even more businesses offering such type of service.

The rapid shift from traditional business models to E-commerce has resulted in significant transformations in supply chain dynamics, freight patterns, and facility requirements at the urban level (Viu-Roig & Alvarez-Palau, 2020; Xiao et al., 2021). The need for quick service requires Q-commerce businesses to locate their delivery hubs as close to customers as possible in order to minimize delivery times and maximize efficiency. This has led to the emergence of a large number of ‘dark stores’ at local or residential neighborhood level. Dark stores are small, micro-fulfillment centers that are designed specifically for Q-commerce (Buldeo Rai et al., 2023; Luthher, 2022; Mukhopadhyay, 2022). They are typically located in densely populated areas and are stocked with a limited variety of high-demand products (Buldeo Rai et al., 2023). The emergence of dark stores will have a significant impact on the planning provisions. For instance, Q-commerce companies will need to ensure that their dark stores are compliant with planning regulations. The emergence of dark stores at the neighborhood level has several planning implications. The location of the dark stores and their compatibility with the land use in which they are located itself is the first implication. The second implication is the spatial requirements of these dark stores such as the number of such facilities at the neighborhood level, area requirements, parking demand, etc. The third is the associated impacts such as the generation of freight traffic at the local level and parking, loading, and unloading issues. The subsequent environmental impacts in terms of noise, emission, and other concerns of safety and security.

However, as this industry is only in its nascent stage, especially in metropolitan cities in developing economies such as Delhi, there is not sufficient research in the area. The development plans typically plan for retail commercial facilities at the neighborhood level designed to receive customers. They are silent on this new typology of logistic facility. This research through a study of the existing Q-commerce industry, especially its last leg, attempts to fill this gap. It would provide valuable information on Q-commerce characteristics, patterns, and impacts that can feed into planning documents. By exploring these dynamics, policymakers and urban planners can make informed decisions to optimize the planning and management of E-commerce-related activities in urban areas.

2. Literature review

This section discusses the concept of Q-commerce and the impacts of E-commerce. While there is a considerable corpus of research addressing the impacts of E-commerce, particularly in the context of urban logistics and last-mile delivery. However, the Indian perspective and the specific implications of Q-commerce remain understudied.

2.1 Genesis of Q-commerce

During the past several years, the emergence of a novel concept known as ‘Quick commerce’ has significantly impacted the E-commerce industry, mostly as a result of the COVID-19 pandemic (Patil et al., 2021). Quick commerce, often known as Q-commerce, has the ability to bring about a significant transformation in the food business by addressing customer demand for everyday items, groceries, vegetables, and fruits. Q-commerce can be defined as a category of E-commerce that involves the speedy delivery of items directly to consumers (Stojanov, 2022). Qcommerce is a business model that focuses on ultra-fast and on-demand delivery of goods and services to customers. It meets the demand for immediate order fulfillment by leveraging technology, logistics, and E-commerce. This approach aims to fulfill customer orders within minutes or hours, primarily in urban or densely populated areas. By utilizing local inventory and delivery resources, Q-commerce providers minimize delivery times, offering same-day or on-demand delivery options (Forbes, 2022). The concept of quick delivery has extended beyond conventional preprepared meals to encompass a wide range of products, including groceries, pharmaceuticals, beauty products, gadgets, and more. This service combines the speed of immediate market transactions with the ease and comfort of remote purchasing (Ranjekar & Roy, 2023). The aspects of convenience, urbanization, and busy lifestyles exert significant influence on customer behavior, prompting companies to prioritize the reduction of delivery time (Wainright, 2021). In cities characterized by high population density and restricted apartment sizes, people often face constraints on available storage space, leading them to adopt a preference for purchasing goods as needed rather than stockpiling. Moreover, the increase in the number of single-member families residing in urban areas has resulted in a corresponding surge in the demand for items that are available in small amounts and necessitate recurring purchases (Gai, 2022). Q-commerce may be considered as the next phase of E-commerce, catering to the customer’s desire for ease and immediate delivery (Chandok, 2021). According to (Villa & Monzón, 2021), customer behavior towards Q-commerce is influenced by several factors, including the opportunity to select from a variety of alternatives, real-time tracking, as well as the convenience of purchasing with just one simple click, all of which are accompanied by instant delivery. Huang & Yen, (2021) distinguished between E-commerce and Q-commerce by considering factors such as customer requirements, time and mode of delivery, and the existence of a store. The delivery timeframe for E-commerce often ranges from one day to one week, however for Q-commerce, it is typically less than one hour. The delivery vehicles used in E-commerce operations exhibit notable distinctions from the two-wheelers commonly employed for Q- commerce deliveries. E-commerce focuses on households with three or more members, whereas Qcommerce is designed for households with a single member. In contrast to E-commerce, where customer behavior is influenced by price sensitivity and discounts, Q-commerce is predominantly led by delivery time sensitivity (Ranjekar & Roy, 2023). The Q-commerce model employs small and scattered dark stores for order processing, in contrast to the utilization of large-scale warehouses in traditional E-commerce operations. Dark stores refer to small-scale stores strategically located in densely populated areas, which play a crucial role in facilitating the quick delivery of products (Forbes, 2022). The key components of the Q-commerce model include dark stores, logistic partners, and technology (Potdukhe et al., 2022).

The COVID-19 outbreak sparked widespread lockdowns that severely restricted public mobility, resulting in a significant increase in consumers’ online purchases, including both deliberate and impromptu buying patterns (Potdukhe et al., 2022). The pandemic had an unprecedented impact on consumer behavior, leading to a rise in demand for Q-commerce deliveries (Elnahla & Neilson, 2021). The rapid expansion of the Q-commerce industry has received attention from many stakeholders in India as well as globally, raising concerns over the long-term viability of Qcommerce enterprises (Ranjekar & Roy, 2023). Furthermore, Q-commerce prioritizes convenience and instant gratification, allowing customers to order products or services whenever needed without planning ahead. It extends beyond retail to include services like food delivery and pharmacy services (Unicommerce, 2023). Efficient logistics and supply chain operations, including advanced routing algorithms and partnerships with local couriers, enable fast and reliable delivery. These characteristics drive the rapid growth of Q-commerce, transforming traditional Ecommerce and shaping customer expectations for speed, convenience, and personalized services. The growth of Qcommerce companies has led to the development and expansion of dark stores in neighborhood areas. These dark stores meet the need for fast delivery and proximity to customers, enabling swift and convenient delivery services. They also facilitate localized inventory and optimize last-mile delivery. The flexibility, scalability, and data-driven decision-making of Q-commerce companies contribute to the expansion of dark stores. As the demand for Qcommerce continues to rise, more dark stores are expected to be developed in neighborhood areas.

2.3 Impact of E-commerce

The volume of literature cites different types of impacts of E-commerce. These can broadly be classified into traffic, environmental, economic, social, technological, and spatial distribution impacts. This section discusses the findings of each of these impacts. The first type of impact is related to traffic and logistics. The explosive growth of Ecommerce has brought about significant traffic and logistics impacts, particularly in the last mile of delivery. According to (Gharehgozli et al., 2017), the rise in e-commerce has resulted in a significant upsurge in the number of delivery vehicles operating within cities. The frequency of last-mile delivery has increased from a few daily trips by local postal services to multiple trips by various logistical providers. The rapid increase in the number of delivery vehicles has a substantial contribution to increased vehicle kilometers travelled (VKT), and traffic congestion especially during peak hours (Hooper, 2019). A study (NHTS, 2019) highlights that E-commerce generates 1 delivery each week per household in US, and 1.35 in New York (NHTS, 2019). Home deliveries in Lyon account for over 130,000 trips including deliveries and shipments every week, or slightly more than 17% of total goods movements in the city (Gardrat et al., 2016). The rapid expansion and development of E-commerce also exert an impact on the mobility of goods. Various elements, such as speed, time, flexibility, information, and cost, exert an influence on the transportation of commodities. Studies (Buldeo Rai, 2019; Janjevic & Winkenbach, 2020; Nguyen et al., 2018).

Sprint Project, (2020) underlined the significance of delivery times in the decision-making process of online shoppers when selecting a delivery method. The occurrence of delivery failure necessitates subsequent delivery attempts, sometimes even extending to a third attempt, and invariably involves the need for logistical processing and communication. These processes incur expenses for logistics providers. According to (Haurillon, 2020) a total of 20 million shipments were subjected to a second delivery in France during the year 2018. On the other hand, the implementation of delivery time service innovations has adverse effects, particularly in relation to the reconfiguration of routes. This is an enormous challenge for logistics service providers, who allocate resources toward the deployment of route management and optimization systems. The adjustments to redirect packages from regular daily rounds to specific evening or weekend rounds, as well as rescheduling rounds to align with specific delivery windows, are factors that contribute to a decrease in delivery efficiency.

The second type of impact is related to the economic implications of e-commerce. Delivery prices remain the most crucial characteristic of any delivery service (Buldeo Rai et al., 2021; Janjevic et al., 2019; Nguyen et al., 2019; Sprint Project, 2020). The notion of ‘free delivery’ specifically promotes regular cost negotiations between online retailers and logistics service providers. The provision of ‘free delivery’ might potentially have negative consequences for both the E-commerce logistics sector and consumer spending patterns, as well as the dynamics of freight flows. This phenomenon encourages irregular purchasing patterns, impulsive buying behaviors, and a high frequency of product returns (Buldeo Rai et al., 2019). Customers are therefore motivated by the offer of complimentary delivery to participate in actions that might lead to inefficient delivery routes, hence exacerbating urban problems. Studies indicate that the alignment between public policies and last-mile logistics systems is not always consistent when evaluating many dimensions of individuals’ quality of life, infrastructure, and the social effect of health risks such as noise and accidents (Buldeo Rai et al., 2017; He et al., 2017).

The third type of impact is related to environmental concerns. Consumer’s irregular and impulsive e-shopping pattern generates a large number of delivery trips per household leading to increased VKT and vehicular emissions. The impact of E-commerce on environmental sustainability varies as a result of the contrasting characteristics of traditional materialistic consumption and online shopping (Ma et al., 2022; Xie et al., 2023). The literature reveals a two-dimensional influence of E-commerce on environmental sustainability. Cheba et al., (2021) found that there is a positive correlation between E-commerce and environmental quality on one aspect, whereas on the other aspect, there is a negative relationship. Proponents of the affirmative correlation between E-commerce and ecological sustainability contend that transportation plays a significant role in the generation of carbon dioxide emissions. According to (Chen & Reklev, 2014), existing literature suggests that a potential strategy for mitigating CO2 emissions is to decrease reliance on vehicles. E-commerce facilitates the execution of company operations without the need for physical travel (Rao et al., 2021). The implementation of remote work and online purchasing has the potential to contribute to the reduction of CO2 emissions, so yielding a favorable impact on the environment (Song et al., 2020). On the other hand, proponents of the adverse relationship between E-commerce and environmental quality contend that the prevalence of online buying in communities is amplified as a result of internet use. Customers have a strong expectation for the punctual delivery of goods, which in turn leads to heightened fuel consumption as a consequence of expedited transportation methods employed by delivery services (Yuan et al., 2022). The rise in fuel usage has a negative impact on environmental sustainability since it leads to an increase in CO2 emissions.

The fourth type of impact is related to technological implications. The utilization of automation technologies exhibits the potential to mitigate labour expenses and augmenting delivery effectiveness, hence potentially restructuring the landscape of E-commerce logistics (Schröder et al., 2018; Touami, 2020). According to (Joerss et al., 2016) it has been suggested that autonomous delivery vehicles will be capable of delivering about 100% of products to customers globally by the year 2025. The influence of E-commerce expands beyond transactional modifications, as it has the capacity to reshape metropolitan environments, modify consumer patterns, and need inventive approaches in logistics and spatial organization. As the pace of technological advancement accelerates, stakeholders must navigate this evolving landscape to ensure a harmonious and efficient urban logistics ecosystem (Joerss et al., 2016).

The fifth and the last type of impact is related to the spatial distribution of e-commerce facilities. The growing popularity of E-commerce has significant implications for the spatial distribution of urban logistics and the patterns of mobility. Traditionally, commercial and business districts were the primary venues for E-commerce activities (Dablanc, 2021). However, with the rise of E-commerce, these activities have become more spread throughout various neighborhoods, including residential areas, resulting in a more diverse and dynamic geography of mobility. The advent of E-commerce has not only brought about transformations in transportation patterns, but has also given rise to novel logistical environments (Rai et al., 2022). Major E-commerce companies have implemented the use of urban warehouses. Urban warehouses have emerged as a novel kind of real estate, necessitated by the growth of Ecommerce (Dablanc, 2021; Hesse, 2004). Two major developments indicate the spatial influence of E-commerce in terms of logistics real estate. Firstly, the establishment of ‘XXL’ fulfillment centers, which are characterized by their substantial size exceeding 50,000 square meters and are commonly referred to as mega-fulfillment centers (Dallas et al., 2022). These centers adhere to the longstanding tradition of locating logistics facilities at a considerable distance from urban hubs. Secondly, there is a parallel pursuit for available space within densely populated urban regions to cater to the escalating requirements of E-commerce deliveries (Dablanc, 2021). The expansion of E-commerce has contributed to the rapid development of urban logistics spaces and logistics micro-hubs. To facilitate the distribution of goods delivery burden and facilitate last-mile deliveries, contemporary models focus on the utilization of small logistical facilities situated in densely populated urban regions.

The challenges and opportunities brought about by E-commerce underscore the need for comprehensive policies and collaborative efforts that integrate logistics into the core of urban planning and development, fostering a symbiotic relationship between E-commerce growth and sustainable urban living. To minimize the number of vehicles, pollution, and pricing, it is imperative to establish a collaborative effort among many stakeholders, including manufacturers, logistics service providers, individuals, and the public sector. Additionally, effective planning processes and rigorous compliance monitoring are essential (Cleophas et al., 2019; Poliak et al., 2022). A grey area that needs greater focus in research is the spatial requirements and/or impacts.

2.3 Bridging the research gap

The existing literature extensively acknowledges the impacts of E-commerce while comparatively less attention has been given to the impacts of Q-commerce. In the context of a developing economy, scant research exists regards the characteristics of Q-commerce, freight patterns, and particular spatial impacts that are not explored extensively. However, there remains a research gap pertaining to the spatial needs and wider consequences of Q-commerce within the Indian urban landscape. In order to bridge this existing knowledge gap, the objective of our research is to study the complex relationship between Q-commerce, spatial planning, and urban sustainability within the context of Indian cities. The objective of this study is to offer practical insights that may be utilized by urban planners, and industry stakeholders by examining the spatial consequences of Q-commerce.

3. Research design

The Q-commerce segment being at a nascent stage in developing economies such as India, there are not enough substantial studies that explore its characteristics, planning requirements, and associated impacts. In this backdrop, this research aims to answer the following research question: what are the key characteristics of Q-commerce in Indian cities; what are its spatial requirements and what are the key impacts? The paper presents a case study that investigates some of the Q-commerce companies operating in Delhi, which is a tier-1 city and the capital of India. Delhi is a major metropolitan area and serves as a significant hub for economic, political, cultural, and commercial activities due to its high level of development. In relation to the rest of the tier 1 cities, Delhi is comparable in terms of its overall development, prominence, and influence. Overall, Delhi, as a tier 1 city, holds a prominent position in terms of its size, influence, and development compared to other cities. According to (Census, 2011), Delhi has the highest population- 46 million, the most registered freight vehicles: 0.24 million (MoRTH, & Govt of India, 2021), and the highest freight traffic- 0.13 million (Kheri et al., 2021). Compared to other tier-1 cities in India, it has the biggest proportion of E-commerce in the warehousing business by transaction volume in the financial years 2021 (61%) and 2022 (41%) (Knight, 2022). Other tier-1 cities in India, also exhibit similar characteristics and hold significant importance in their respective regions and as such the findings of this study can be generalized to other tier-1 cities.

The study specifically examines the grocery segment within the E-commerce category in one of Delhi’s planning zones. This particular focus is due to the rapid growth observed in the grocery and food delivery sector within the Qcommerce segment (IBEF, 2022). Unlike food deliveries, which can be facilitated by partnering with local restaurants, the grocery segment has distinct spatial requirements, and hence this particular segment was selected for this study.

Four different Q- commerce companies were selected namely Blinkit, Amazon Fresh, Flipkart Quick, and Swiggy Instamart. This study employed a mixed-methods approach to research. The survey consists of semi-structured interviews with delivery personnel and local warehouse managers in the selected zone of Delhi. The information collected through the surveys includes details of trip characteristics such as the number of trips, trip distance, trip time, mode of delivery, and total delivery staff, while information regarding spatial characteristics such as location and size of the store, land use, accessibility, parking space, unloading area, etc. In addition to mapping the local distribution hubs namely dark stores in the study communities, whole-day delivery patterns of delivery personnels were documented. Different sample sizes for store selection, warehouse managers, and delivery respondents were considered for different companies. A sample of 30% Blinkit stores was selected i.e. 4 out of 13 stores were surveyed, while for the other three companies i.e. Amazon Fresh, Flipkart Quick, and Swiggy Instamart, there was only single store per company in the study area, so 100% sample was considered. For the delivery persons survey, a sample of 10% was taken for all four Q-commerce companies. A total of 24 delivery persons were surveyed. The analysis includes profiling the key characteristics of the industry such as its operational model, catchment area, supply chain, and the trip characteristics of the last leg of delivery.

4. Case study profile

4.1 About the study area

The study focuses on one of the zones of Delhi. Delhi is divided into 18 zones, out of which Zone-E i.e. East Delhi has been selected for the purpose of the case study. East Delhi is situated on the eastern banks of the river Yamuna which divides Delhi into two parts. East Delhi has an area of 64 square kilometers, with a population of 0.17 million as per 2011 Census data. Zone E has been selected for three major reasons. Firstly, Zone E emerges as the second most populous and has the second-highest population density among all the planning zones in Delhi. Secondly, Zones E, F, and G exhibit the highest concentration of Q-commerce dark stores when compared to the other zones. Particularly, Zone E and Zone F have the highest dark store density among the other zones. Thirdly, Zone E, which is located on the extreme eastern edge of Delhi, shares an administrative boundary with the neighboring state of Uttar Pradesh. This distinctive geographical location attracts a variety of diverse populations, each with distinct socioeconomic characteristics. Figure 1 highlights the geographical location of the study area within Delhi.

Figure 1. Study area profile.

4.2 Q-commerce companies profile

4.2.1 Delivery time

One of the factors that distinguish the four Q-commerce companies is their delivery time. Amazon fresh services provide a scheduled delivery of 2 hours, while Flipkart Quick delivers within 45 to 60 minutes. Both companies carry chain trips and have a bigger serving area. On the other side, Swiggy Instamart and Blinkit are dark stores, which are located very close to residential areas, with a very small delivery time window. Swiggy Instamart has two delivery models i.e. ‘Deliver now’ and ‘Deliver Later’. ‘Deliver now’ provides an instant delivery within 15-30 minutes, while the ‘Deliver Later’ option provides deliveries in their scheduled delivery timings. Blinkit has also established its dark stores close to the residential area and provides instant delivery within 10 minutes. Due to its short delivery time, it delivers one order at a time only.

4.2.2 Catchment area

The catchment area is another key factor that distinguishes the four Q-commerce companies. Since Amazon fresh and Flipkart Quick have a higher delivery time, they are serving larger areas and have divided areas into pincodes. Amazon has a single store (located in North Delhi, outside the study area but serving the study area), while Flipkart has two stores located in the study area, which serves most of the East Delhi areas. Flipkart Quick stores serves the study area through two different stores, both lying inside the study area. Blinkit through its 14 stores in the study region serve an average radius of 2 to 3 kilometers, while Swiggy Instamart has 5 stores in the study area serving an average radius of 3 to 5 kilometers. Table 1 describes the general characteristic of Q-Commerce companies.

Table 1. General characteristics of Q-Commerce companies.

Company Name	Delivery Time	Catchment Area
Amazon Fresh	120 minutes	8–10 Kilometers
Flipkart Quick	45-60minutes	8–10 Kilometers
Swiggy Instamart	15–30 minutes	3–5 Kilometers
Blinkit	10minutes	2–3 Kilometers

4.2.3 Supply chain

Blinkit has established its dark stores within the local areas, to serve quickly. Groceries and F&V (Fruit and Vegetables) products are the two categories of goods delivered to Blinkit Stores. Both commodities have their different supply chain. All grocery products are processed at a fulfillment center located at a distance of 30 kms in a fringe area, where numerous brands deliver goods from their manufacturing hubs. While F&V products are handled at other urban areas 38 kms, where they are received from wholesale vegetable markets located at Azadpur mandi, and some nearby rural areas also. Unlike Blinkit, Swiggy does not own any warehouse. It has tied up with various 3PLs (Third-party logistics) for different items, such as for the supply of frozen goods located in a fringe area at 42 kms, for the supply of F&V products located in other rural areas via intermediary hubs at 35 kms, for the supply of other goods which are directly supplied from individual brands/vendors such as Mother Dairy, Licious, etc. The goods are supplied to the distribution stores during the night from 12 am to 3 am. The goods are transported via LCVs at all establishments. Both Amazon Fresh and Flipkart Quick have partnered with various 3PL for their regular supply and distribution of goods. Amazon Fresh imports its grocery products from its own warehouse established in a regional area, both F&V products and frozen items are imported from fringe areas, while Flipkart Quick imports its grocery products from a 3PL warehouse located in other urban areas and various other products from different merchants located across NCR, Haryana and Uttar Pradesh. Figures (2-5) describes the schematic supply chain of each Q-commerce company.

Figure 2. Supply chain of Blinkit.

Figure 3. Supply chain of Swiggy Instamart.

Figure 4. Supply chain of Amazon Fresh.

Figure 5. Supply chain of Flipkart Quick.

5. Analysis

The section is divided into two subsections. The initial section offers insights into the trip characteristics of each Qcommerce delivery company, such as average travel length, average trip time, mode share, and vehicle capacity. The subsequent section investigates different impacts associated with Q-commerce delivery operations.

5.1 Trip characteristics

The focus of this paper is only on the last leg (i.e. Leg-1) of the supply chain which is from delivery stores to consumers as described in Figure 6 since they impact the local neighborhood level environment as well as the urban freight the most. Table 2 summarizes the trip characteristics of all four E-commerce companies.

Figure 6. Legs of supply chain.

Table 2. Trip characteristics of E-commerce companies for leg-1.

Company Name	Avg. Trip Distance (Kilometres)	Avg. Trip Time (minutes)	Mode Share	Vehicle Carrying Capacity(Kilogram)
Blinkit	2.33	9	100%: 2-Wheeler	15
Swiggy Instamart	5.85	20	100%: 2-Wheeler	15
Flipkart Quick	23.8	63	100%: 2-Wheeler	15
Amazon Fresh	14.6	51	100%: 2-Wheeler	36

5.1.1 Average trip distance

The average trip distance for leg-1 was calculated based on averaging 3 trips of each sample delivery person (total of

26). For the 10–30 minutes range companies (i.e. Blinkit & Swiggy Instmart) the average trip distance of leg-2 was 170 km, While the average trip distances of leg-1 for Blinkit, Swiggy Instamart, Flipkart quick, and Amazon Fresh came out as 1.8, 7.25, 23.8, and 14.6 kilometers respectively. The differences in average trip length can be explained through the differences in catchment area discussed in section 3.2.2 and the supply chain of each of these companies, explained previously in Figures 2 to 5.

5.1.2 Average trip time

A similar procedure is used to determine the average time taken by a delivery person to deliver goods from each distribution store/hub to the consumer. The average trip time for Blinkit, Swiggy Instamart, Flipkart quick, and Amazon Fresh came out as 9, 20, 63, and 51 minutes respectively.

5.1.3 Mode share

In order to deliver goods on time in busy neighborhoods with narrow streets, all four companies use only two-wheelers for delivery trips. As a result, two-wheelers account for the entire mode share. Each of the four firms has set a maximum weight limit for a delivery person’s consignment, which is 36 kg for Amazon Fresh and 15 kg for Flipkart Quick, Swiggy Instamart, and Blink it. The consignment gets divided if the order weight exceeds the maximum limit. The mode share of leg-2 distribution is 50% each for LCV and HCV for Binkit while it is 100% LCV for Swiggy Instamart.

5.2 Impacts

The section provides insights into various kinds of impacts associated with last-mile Q-commerce operations. The impacts have been broadly classified into spatial, traffic, and environmental Impacts. The focus of the paper is on the spatial impacts and hence different spatial parameters are described extensively.

5.2.1 Spatial impacts

In this section, a comprehensive analysis of key spatial parameters for E-commerce delivery stores is described. Table 3 provides insights into the allocated land use, existing land use, surrounding land use, access road width, service road width (where applicable), built-up area, and existing on street two-wheeler parking. This examination sheds light on the intricate interplay between spatial planning and the evolving demands of modern E-commerce operations.

Table 3. Spatial parameters for E-commerce delivery stores.

Delivery Store	Allocated Land use	Observed Land use	Access Road widths (meters)	If access via service road (Yes/No)	Built-up Area (square meter)	On-street parking (maximum observed)
Store-1	R	R	25	Y	280	16
Store-2	I	I	15	N	240	18
Store-3	C	C	50	N	280	15
Store-4	T	A	12	N	180	13
Store-5	P	P	45	Y	215	16
Store-6	R	C	25	N	185	14
Store-7	A	M	52	N	250	17
Store-8	R	M	12	N	220	15
Store-9	R	M	22	N	180	16
Store-10	C	C	20	N	290	21
Store-11	R	M	10	N	300	19
Store-12	R	M	22	N	180	14
Store-13	R	R	10	N	270	12
Store-14	R	R	6	N	240	14
Store-15	R	R	42	Y	260	15
Store-16	R	C	12	N	285	13
Store-17	R	R	13	N	160	12
Store-18	R	M	22	N	205	17
Store-19	C	C	42	Y	300	15
Store-20	C	C	45	Y	290	17
Store-21	R	M	18	N	180	16
Store-22	R	M	15	N	165	18
Store-23	REC	C	55	Y	372	37
Store-24	I	P	30	Y	665	65

A- Agricultural; C- Commercial; I- Industrial; M- Mixed; P- Public & Semi-Public; R- Residential; REC- Recreational; T-Transportation; Y-Yes; N- No

A. Land use

An analysis was carried out on the impact of delivery stores on the land use pattern of E-commerce delivery stores. Land use conformity and compatibility are essential factors to consider when evaluating the impact of delivery stores on an area’s land use pattern. Conformity refers to the degree to which a land use adheres to the designated zoning regulations and master planning guidelines. Compatibility, on the other hand, relates to the ability of different land uses to coexist harmoniously within the same vicinity. It was found that stores were located on different land uses such as commercial, industrial, residential, agricultural, recreational, transportation, and public and semi-public land use. The diverse distribution of delivery stores necessitates an assessment of both land use conformity and compatibility to understand their implications on urban development.

A comparison was made between the existing land use and the allotted land use as per the master planning regulations set by the Delhi Development Authority (DDA, 2021). The comparison provides insights into the degree of conformity of the delivery stores with the designated land use categories. As per allocated land use, it was analyzed that 58% (14) of the Q-commerce delivery stores located, were established on residential land use, indicating a significant presence of non-conformity with the designated land use categories. Furthermore, 16% (4) of the stores were established on commercial land use and the rest of the stores were located on different land use as shown in Figure 7. It was also found that around 66% of the Q-commerce delivery stores were located in non-compatible land uses other than commercial and industrial land use. The majority of the stores are located in residential areas could be the reason for their small delivery time, proximity to customers, customer convenience, delivery efficiency, and low delivery cost. The violations in land use from commercial to residential could be due to the reason that commercial real-estate businesses operating in residential areas are more affordable than in prime commercial zones, making cost-effective decisions for Q-commerce companies. As some of the stores were also located on commercial and industrial land use, is due to the reason that the available space in any designated land use offers a more competitive advantage and also caters to a particular demand zone. As a part of semi-structured interviews with store managers, it was found that some of the stores were about to shout down because of reasons that are directly related to planning such as unavailability of affordable commercial space, land use violation, and low demand zones due to socio-economic characteristics of the catchment area.

Figure 7. Land use variation of Q-commerce delivery stores.

A review of planning documents (AMC 2006; AUDA, 2018; BDA, 2017; CEPT, 2014; CEPT, & UMTC, 2017; CMDA, 2019; CIDCML, 2014; DTCP, 2010; DULT, 2019; GHK, 2007; HMDA, 2013; HUDA, 2008; HMDA, 2013; JDA, 2010; JDA, 2006; JPS Associates, 2006; KMA, 2008; LMC, 2015; Mehta & DIMTS, 2012; MCGM, 2016; PMC, 2008, 2012; PMRDA, 2021; SUDA, 2017; TCPO, 2020; TPCO, 2012; UMTC, 2010, 2016; VUDA, 2019), encompassing master plans, city development plans, and comprehensive mobility plans from various tier-1 and tier-2 cities in India was conducted to ascertain the prevailing land use regulations concerning E-commerce delivery stores. The analysis revealed that none of the examined planning documents explicitly cater to the land use provisions for such types of E-commerce facilities. Mixed land use regulations, development control norms, and land use conversion policies are designed to enable land to be used for purposes other than its original intended use. These policies establish the framework for when and how such changes can be implemented in various scenarios. Due to the ambiguity in planning regulations and the absence of standardized rules for E-commerce operations, these facilities emerge across diverse land uses, often affecting the surrounding environment. Development authorities might take time to catch up with emerging trends. The time required to draft, review, and implement new regulations can result in a lag between the emergence of new concepts and the establishment of relevant policies. In accordance with this, similar studies conducted in a variety of global cities highlight the significance of effective spatial planning for E-commerce facilities. Similar studies conducted in Paris (Buldeo Rai et al., 2023; Dablanc, 2023), New York (Schorung et al., 2023), Hong Kong (Xiao et al., 2021), London (Allen et al., 2021) highlight the multifaceted challenges posed by the rapid growth of Ecommerce and its associated infrastructure. As the global E-commerce landscape continues to evolve, proactive spatial planning emerges as a key factor in assuring the coexistence of efficient E-commerce logistics and the preservation of urban functionality and livability.

B. Built-up area

It was observed that all the stores were located on the ground floor of a building or establishment. The built-up area of each of the stores was calculated, and it was found that the dark store has a relatively close range of built-up area from 160 to 300 square meters. The average built-up area of dark stores was estimated to be 235 square meters. The distribution of stores based on built-up area is shown in Figure 8. It was estimated that a unit built up area of darks store generates 2.15 consignments per day per store, while delivery stores with greater than 30 minutes delivery time generates 1 consignment per day per store. Due to the nature of dark stores established on smaller area of land, along with time window constraints tend to handle a higher volume of consignments per unit of space compared to the delivery stores with longer delivery times, which handle fewer deliveries. Other parameters such as delivery cost, storage size, route network algorithms, etc. must be considered for effective operational planning and resource allocation.

Figure 8. Distribution of Q-commerce stores.

C. Parking requirement

The emergence of E-commerce delivery stores and Q-commerce dark stores in urban areas has introduced a novel aspect to the difficulties and requirements associated with parking. The growth of E-commerce and on-demand services has resulted in a significant increase in the number of delivery vehicles operating through urban areas that are currently impacted by traffic congestion. The availability of sufficient parking spaces in close proximity to these companies is crucial for ensuring uninterrupted operations. However, there are significant challenges due to the lack of available parking locations, strict restrictions, and a limited supply of spaces along the curb. Due to the rise in Ecommerce, there are now significantly more delivery vehicles using city streets for parking, loading, and unloading. While this tendency gives customers convenience, it presents a number of problems for the nearby urban environment.

Congestion on already congested streets is one of the main problems with E-commerce last-mile delivery parking. Delivery vehicles frequently double park or occupy curbside spaces for long periods of time, obstructing traffic flow and resulting in traffic congestion.

To determine the parking requirement of E-commerce last-mile delivery stores, we conducted a parking survey at each of the delivery stores, estimated the number of vehicles parked during peak hours, and compared the results to parking standards established by the planning authorities. The majority of the delivery stores are dark stores in the study area. The parking requirement and the total built-up area of each dark store were calculated. According to the parking standards for commercial-retail land use, 1 equivalent car space (ECS) is required to be provided for a built-up area of 50 square meters in Delhi. From our primary survey analysis, it was found that a dark store, with a built-up area of around 235 square meters, requires a minimum of 16 two-wheeler parking, while the parking standard recommends parking space for 10 two-wheelers, which is slightly less than the existing requirement. As the demand for Qcommerce increases, more parking will be required. Since the current parking demand exceeds the parking standards, the standards must be modified to accommodate the current E-commerce market penetration. According to our analysis, a dark store with retail-commercial land use requires 1 ECS per 30 square meters of total built-up area. This modifies parking requirements to consider the shifting dynamics of the E-commerce environment. Urban transport planners and E-commerce companies must collaborate to implement strategic solutions to such issues in the city. The provision of loading and offloading zones, particularly for delivery vehicles, can reduce congestion and improve traffic flow. To reduce the traffic congestion on urban streets during peak hours, E-commerce companies can consider rescheduling deliveries during off-peak hours. The issues with parking brought on by last-mile E-commerce delivery operations have significant effects on the urban environment. The occurrence of traffic congestion, pollution, limited parking options, and disrupted pedestrian pathways pose significant challenges that require the collective involvement of urban planners, businesses, and local authorities. It is imperative to foster collaboration among these stakeholders to devise sustainable solutions that prioritize efficiency, environmental preservation, and the overall welfare of urban communities.

D. Access road widths

The analysis finds a significant variance in access road width between store locations. This variance is due to the various land uses where the stores are located. It highlights the importance of adaptable spatial planning solutions that take into consideration the distinct qualities and restrictions of each location, particularly in terms of access road width. The distribution of stores was observed throughout different road widths. It was observed that among all the stores, only 30 % of the stores had access via the service road. Service roads are generally present on the higher hierarchy of the Indian road network as access points to properties. Therefore, the specific placement of such stores along these roads does not pose a significant concern. The analysis of store distribution across various access road width categories yields significant information as shown in Figure 9. It was found that around 17% of the stores have access roads that are less than 10 meters wide, suggesting potential challenges in accommodating freight traffic and parking requirements. Furthermore, around 29% of stores are located along access roads 11 to 20 meters wide, which offers a somewhat favorable environment for freight operations. The distribution across differing width ranges, including 21 to 30 meters (25 %), 41 to 50 meters (21 %), and greater than 50 meters (8 %), highlights the need for flexible spatial planning strategies to accommodate diverse operational needs. The section emphasizes the significance of access road width can be from a logistical perspective for Q-commerce facilities. The evidence shows that these establishments’ placement along various access roads with different widths has an impact on how efficiently they operate and handle freight traffic. The varied distribution of stores across various kinds of access roads highlights the significance of developing spatial planning solutions to accommodate the distinctive characteristics of each location while ensuring efficient freight operations and logistical functionality.

Figure 9. Classification of stores by width of access road.

5.2.1 Traffic and environmental impacts

It is estimated that around 0.01 million vehicle kilometers traveled (VKT) is generated from the last leg of the supply chain from all the 4 case study companies within the study area. The VKT is calculated from the following formula:

VKT (Vehicle Kilometers Travelled) = (Total delivery person per store) x (Total delivery person per store) x

(Average trip length) x (Total no. of stores)

An analysis was conducted to assess the freight pattern for a given built-up area and the number of delivery stores. It was calculated for the two types of delivery facilities. It was estimated that on an average a dark store generates 16 kms of VKT per day per store, while delivery stores with delivery time greater than 30 minutes generates 15.8 kms of VKT per day per store. It is analyzed that within a given area, dark stores being smaller in size (area) and lager in quantity possesses same amount VKT as the other type of delivery store, which is being fewer in number and larger in area. It was also found that on an average, each dark store generates 3867kms of VKT per day, while stores with higher delivery time generate 8199 kms of VKT per day. It signifies that a significant freight trip are being generated by each facility which underscores substantial contribution to local traffic congestion and environmental emissions.

Traffic impacts are interlinked with environmental impacts, as high VKT generated from facility will lead to subsequent environmental implication. The emission generated form each stores is estimated and it was found that the total estimated emissions is approximately 125 kgs/day (CO2 equivalent), is generated collectively by all four Qcommerce companies. The emission load factors recommended by the (Institute of Urban Transport, 2014), for several pollutants, including PM_2.5, NO_X, CO, and VOC, particularly to delivery vehicles, i.e. two-wheelers, were used in this computation. The average emissions generated by each store were estimated based on the analysis of primary survey data. It was found that stores, on average, contribute approximately 8.23 kilograms (kgs) of emissions. However, a significant difference was observed when considering companies with varying delivery times. Companies with delivery times ranging from 10 to 30 minutes were found to have significantly lower emissions, averaging around 6.38 kgs per store. On the other hand, companies with delivery times exceeding 30 minutes had considerably higher emissions, with an average of approximately 10 kgs per store. It was analyzed that companies capable of achieving shorter delivery times demonstrate higher emission efficiency when compared to those with longer delivery times. It strongly suggests that companies with faster delivery times exhibit higher overall emission efficiency, resulting in a significant reduction in environmental effects. Table 4 shows the total emissions generated by each company.

Table 4. Estimated VKT and C02 equivalent emissions.

Company Name	Delivery Staff	Total Trips	ATL (Km)	No. of Stores	VKT (Km)	PM 2.5 (gm/km)	NOx (gm/km)	CO (gm/km)	VOC (gm/km)	Emissio n (kg)	Emission/Store (kg)
Blinkit	18	33	2.33	17	47057	1835.21	2164.60	35904.24	17975.65	58	3.41
Swiggy	25	26	5.85	5	38025	1482.97	1749.15	29013.07	14525.55	47	9.4
Amazon Fresh	65	4	14.6	1	7592	296.08	349.23	5792.69	2900.14	9	9.34
Flipkart Quick	37	5	23.8	1	8806	343.43	405.07	6718.97	3363.892	11	10.83
Average Emissions/Store (Kg):											8.23

Note- Emission factors: PM2.5–0.039 gm/km, NOx- 0.046 gm/km, CO- 0.763 gm/km, VOC-0.382 gm/km

6. Key findings

The study revolves around the multifaceted impacts of last-mile Q-commerce operations, focusing on spatial, traffic, and environmental dimensions. In the realm of spatial impacts, the analysis uncovers a diverse landscape where delivery stores span across various land use categories, including commercial, industrial, residential, agricultural, recreational, transportation, and public and semi-public land use. Notably, a significant portion of Q-commerce stores is located in non-conforming residential areas, reflecting proximity advantages and cost-effective decisions. However, in some cases, when commercial land use is of a higher hierarchy or has high rental property rates, it becomes unaffordable to these companies. The study also finds that these stores shut down because of reasons that are directly related to planning such as unavailability of affordable commercial space, land use violation, and low demand zones due to socio-economic characteristics of the catchment area. A lot of freight traffic is generated from these stores within the local area, due to the movement of the last mile delivery vehicles and heavier goods vehicles from distribution hubs to delivery stores, occurring throughout the day, as a result of which, leads to negative externalities such as traffic congestion, noise, and emissions on the local as well as collector streets within neighborhood areas. In the absence of adequate planned spaces provided through planning documents, such facilities either come up in unplanned areas or in areas that are not compatible with surrounding land use, as evidenced in this study. The analysis of built-up areas highlights the characteristics of dark stores predominantly located on the ground floor, boasting a relatively consistent built-up area range of 160 to 300 square meters, serving a radius of 3 kms. The estimated average built-up area of dark stores is approximately 235 square meters. Studies conducted globally (Mariquivoi, 2022; Matthieu, 2023; Paché, 2022b) revealed similar operational characteristics in Q-commerce delivery stores, and as such these stores are located at the ground floor of a building, covering a built-up area varying from 100 to 300 square meters and serve a radius of 2 kilometers. It was found that parking requirements exceed existing standards, necessitating revisions to accommodate the growing E-commerce market. Access road width variations emphasize the adaptability required in spatial planning strategies. Traffic and environmental impacts indicate significant VKT generation and emissions, with delivery times playing a crucial role in emission efficiency.

The study raises several issues that should be taken into account when preparing planning documents and providing freight and logistics spaces. Some of these issues include stores located on non-compatible land uses, frequent movement of delivery fleet drivers throughout the day creating inconvenience to local residents, and neighboring property owners as well as hindrance to the road user. Studies conducted globally highlight some similar issues such as stores located at the ground floor of buildings and in a densely populated area (Matthieu, 2023), incompatible land use (Rai et al., 2023), non-uniform spatial pattern of delivery stores (Dablanc et al., 2022), fuelbased two-wheeler used for last-mile delivery (Mariquivoi, 2022), on-street parking and occupying public space (Paché, 2022b), occurrence of various freight activities throughout the day at local neighborhood level areas (Matthieu, 2023; Paché, 2022a). The study raises questions regarding the permissibility of non-visitorcentric E-commerce stores in retail commercials, the feasibility of establishing them in MRT station precincts, and the need for a hierarchy of facilities for E-commerce activities in urban areas. Moreover, issues such as minimum abutting street width and parking space availability for store locations must be addressed to prevent hindrances to road users. Stores located in residential or incompatible land-use areas have profound impacts on surrounding neighborhoods, necessitating corrective frameworks for freight vehicle enumeration, and freight vehicle movements at the neighborhood level. In conclusion, this research provides valuable guidance for policymakers and stakeholders, offering insights that prioritize operational efficiency, environmental stewardship, and the well-being of urban communities in the face of the growing influence of Q-commerce operations.

7. Conclusion

In conclusion, the study has explored spatial, traffic, and environmental impacts of last-mile Q-commerce operations which are interlinked with each other. The spatial analysis reveals the diverse land use patterns, non-conformity with spatial planning regulations, parking issues, and the importance of access road width in efficient operations. The traffic and environmental analysis highlight the significant contribution of Q-commerce delivery facilities to traffic congestion and emissions, with operational parameters like delivery time frame playing are particularly important for emission efficiency. As the global E-commerce landscape continues to evolve, proactive spatial planning emerges as a crucial factor in ensuring the coexistence of effective E-commerce logistics and the advancement of urban functionality and livability. Collaboration between urban transport planners, local authorities, and logistics operators is crucial to address these challenges effectively. This study provides valuable insights for city planners, policymakers, and E-commerce logistic operators firms to develop sustainable solutions by prioritizing efficiency, sustainability, and the well-being of urban communities in the face of expanding Q-commerce operations.

Since this paper highlights findings based on a pilot study, future research would involve a scaling up of the same study with a larger scope with respect to area as well as other parameters. Other parameters such as distribution of different hierarchies of facilities, freight flow, demand zones, stakeholder engagement, infrastructure capacity, etc. can be considered within the same study. Also, a consumer perspective to model the overall demand generation and online shopping behavior for Q-commerce would further enrich the study. Within the Q-commerce segment, different models exhibit different characteristics and subsequent impacts as also evidenced in this study. Future areas of research can examine which type of operational model is more efficient and sustainable in last-mile E-commerce delivery.

Disclosure statement

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