Crowdsourcing Last Mile Delivery: Strategic Implications and Future Research Directions

AuthorVincent E. Castillo,John E. Bell,William J. Rose,Alexandre M. Rodrigues
Publication Date01 March 2018
Date01 March 2018
Crowdsourcing Last Mile Delivery: Strategic Implications and
Future Research Directions
Vincent E. Castillo
, John E. Bell
, William J. Rose
, and Alexandre M. Rodrigues
University of Tennessee
University College Dublin
The rise of e-commerce over the past 20 years has created an increased need for responsive omnichannel distribution to meet the last mile
challenge. Some companies are experimenting with the use of the sharing economy business model to augment distribution strategies. The
use of so-called Crowdsourced Logistics(CSL) is becoming more prevalent in practice, but the role in logistics strategy of this new phe-
nomenon has not been thoroughly investigated and understood. Using a contingency theory lens, this research contributes a nascent understand-
ing of how CSL performs in terms of logistics effectiveness by simulating same-day delivery services from a distribution center to 1,000
customer locations throughout New York City under dynamic market conditions and by comparing the results to those of a traditional dedicated
eet of delivery drivers. The ndings are analyzed to suggest how rms may nd strategic benet using CSL. An agenda for future research is
provided to explore these strategic implications and to deepen knowledge about the CSL phenomenon.
Keywords: crowdsourcing; urban logistics; omnichannel distribution; transportation; supply chain management
Over the past 20 years, the rapid growth of e-commerce has led
to an evolution in supply chain management strategy and prac-
tice (Brynjolfsson et al. 2013; Bell et al. 2014; Peinkofer et al.
2015; Ta et al. 2015). Customers increasingly require anytime,
anywhere demand fulllment, necessitating improved inventory
management and distribution strategies (Napolitano 2013). In
response, companies seek to integrate innovative transportation
technologies into existing distribution systems. One such innova-
tion emerges from the sharing economyclass of business mod-
els, offering multiple users temporary asset ownership benets at
a reduced cost (Howe 2006; Lamberton and Rose 2012; Miller
2013). One of the most popular models is ridesharing, facilitated
by companies such as
Uber and Lyft, which distribute costs and
benets by connecting independent car owners and passengers
via a mobile or computer application. Large rms, including
Amazon and UPS, are increasingly investing in adaptations of
the ridesharing service model to perform same-day delivery ser-
vices, a phenomenon colloquially known as Crowdsourced
Logistics(CSL) (AmazonFlex 2016; Savelsbergh and Van
Woensel 2016; Supply Chain 24/7 2016). In the CSL business
model, a shipper procures transportation services via a mobile or
computer application directly from members of the crowd who
provide those services as an independent contractor using a per-
sonally owned vehicle asset.
CSL relates to several emerging areas in supply chain research.
For example, omnichannel distribution research lends insight into
how rms simultaneously manage in-store and online channels to
create customer value (Neslin et al. 2006; Verhoef et al. 2015).
The quality of physical distribution service (Rabinovich and Bai-
ley 2004; Rabinovich et al. 2008) and effective returns manage-
ment can improve online retailer performance (Grifs et al.
2012a; Rao et al. 2014). Additionally, reductions in order fulll-
ment cycle time enhance customer referral behavior, thus leading
to increased rm performance (Grifs et al. 2012b). Common
across these research efforts is the examination of logistics effec-
tiveness in an e-commerce context. CSL relates to these research
streams as a transportation mode within the last mile logistics
strategy, but one whose impact on logistics effectiveness has not
been fully examined in supply chain literature. Thus, the goal of
this research was to compare CSLs effectiveness as a last mile
delivery mode to that of traditional dedicated delivery modes. In
pursuit of this goal, the following research question is asked: How
does a crowdsourced eet compare to a traditional dedicated
courier eet in terms of logistics effectiveness under dynamic task
environment conditions?
Related academic disciplines provide relevant work for initiat-
ing this research effort. Operations researchers have compared
owned and outsourced transportation assets (Hoff et al. 2010),
but this research tends to be concerned with cost minimization or
eet mix optimization rather than logistics effectiveness
(Saunders et al. 2015). Additionally, most related vehicle routing
research has not explored the effect of uncertainty in the supply
of drivers, which characterizes CSL (Eksioglu et al. 2009;
Lahyani et al. 2015). Furthermore, previous logistics research on
same-day delivery services has demonstrated that task environ-
ment conditions, such as delivery windows and demand uctua-
tions, affect performance (Campbell and Savelsbergh 2005;
Boyer et al. 2009). However, this research also does not explore
these task environment conditionsimpact alongside a resource
constraint (such as vehicle supply uncertainty) on strategy and
performance (Autry et al. 2008). Thus, there is a gap in aca-
demic knowledge with respect to logistics effectiveness associ-
ated with dynamic market conditions and the uncertain resource
supply present in the sharing economy (Hossain and Kauranen
2015). As a result, the current research uses contingency theory
Corresponding author:
Vincent E. Castillo, Department of Marketing & Supply Chain
Management, University of Tennessee, 305 Stokely Management
Center, 916 Volunteer Boulevard, Knoxville, TN 37996, USA;
Journal of Business Logistics, 2018, 39(1): 725 doi: 10.1111/jbl.12173
© Council of Supply Chain Management Professionals
(Drazin and Van de Ven 1985) to connect logistics strategy with
task environment conditions and uncertainty in a supply of logis-
tics assets (Autry et al. 2008).
Examining CSL for same-day delivery services under various
environmental conditions answers Hossain and Kauranens
(2015) call for understanding crowdsourcings strategic implica-
tions by comparing potential benets and risks. Crowdsourcing
provides a quick means of performing deliveries as drivers are
independent contractors using personally owned vehicles to pro-
vide logistics services. However, as crowdsourced drivers man-
age their own schedules, CSL increases uncertainty relative to
more stable dedicated vehicle eets with known capacities and
availabilities (Karger et al. 2011; Ndubisi et al. 2016). This form
of resource sharing, or collaborative consumption, distributes
costs and benets across multiple users (Belk 2014; Cohen and
Kietzmann 2014), but the crowd (i.e., specic crowd members)
chooses whether or not to provide the rm with a strategic
resource (Daft et al. 1988). This makes the crowd both an
uncontrollable environmental factor and a potential structural
resource, thus increasing uncertainty and risk (Drazin and Van
de Ven 1985; Daft et al. 1988). As collaborative consumption
grows more popular among consumers (Matzler et al. 2015), the
efcacy of crowdsourced services, such as logistics, should be
explored more thoroughly to understand how they can contribute
to customer value (Hossain and Kauranen 2015) and how certain
task environment conditions affect the creation of customer value
(Venkatraman 1989).
In exploring CSLs impact on logistics effectiveness under
certain task environment conditions, we make three contributions
to supply chain literature. First, we develop a systems-level
understanding of the CSL phenomenon as a component of a
rms last mile distribution strategy. Second, we suggest how
CSL can be leveraged strategically by comparing logistics effec-
tiveness of a crowdsourced eet of delivery drivers characterized
by high vehicle supply uncertainty to that of a dedicated eet of
drivers with low vehicle supply uncertainty. Finally, we present
a future research agenda to stimulate further investigation of the
CSL phenomenon.
To make these contributions, we perform a stochastic discrete
event simulation model informed by secondary data and discus-
sions with managers from courier companies in major American
cities (Bowersox and Closs 1989; Goldsby et al. 2006). Drawing
upon previous research on both courier (Gendreau et al. 2006;
Van Hentenryck and Bent 2006) and same-day delivery services
(Campbell and Savelsbergh 2005; Boyer et al. 2009), we intro-
duce vehicle supply uncertainty to compare CSLs performance
with that of a traditional, dedicated eet and to assess how the
organizational task environment affects this relationship. We also
conduct an exploratory post hoc analysis to further explore the
relationship between the uncertainty associated with a supply of
crowdsourced drivers and logistics effectiveness.
The remainder of this article briey reviews previous research
relevant to this study, which is followed by hypotheses develop-
ment. The simulation model development process (SMDP) is
then described, followed by the exploratory post hoc analysis. A
discussion of the simulations results and implications for theory
and practice follow. Finally, we present a future research agenda
for improving understanding of CSL for last mile distribution.
This study of the CSL phenomenon can be informed by literature
in the last mile logistics, transportation brokerage, crowd-
sourcing, and vehicle routing research streams. Scholars have
been considering last mile transportations importance in distribu-
tion strategies since e-commerces initial rise to prominence in
the late 1990s and early 2000s. Bridging the last mile is consid-
ered critical to the online shopping experience and to developing
effective distribution strategies (Lee and Whang 2001; Esper
et al. 2003; Boyer and Hult 2005; Kull et al. 2007; Boyer et al.
2009). More recently, scholars have examined the state of
omnichannel management (e.g., Herhausen et al. 2015; H
et al. 2016; Ishfaq et al. 2016; Mena and Bourlakis 2016), which
encompasses last mile transportation; however, CSLs role in
such strategies has yet to be explored.
Because CSL is enabled through creating electronic exchange
markets, literature on transportation brokerage also provides refer-
ence points for how scholars may think about the CSL phe-
nomenon. Electronic transportation markets (ETMs) facilitate
transactions between buyers and sellers of transportation services,
resulting in lower information-seeking, bargaining, and policing/
enforcement costs (Beilock and Shell 1992; Goldsby and Eckert
2003). Companies that create mobile or computer-based applica-
tions to connect buyers and sellers of transportation services (i.e.,
crowd members), such as Postmates or Deliv, act as transporta-
tion brokers, providing similar benets to those of ETMs for last
mile delivery in exchange for fees. Like most transportation bro-
kerage rms, the creators of applications for CSL typically do not
have many assets (Ashenbaum et al. 2012); but a main difference
of CSL applications is the supply chain tier where the purchased
transportation is provided. Transportation brokerage rms typi-
cally focus on upstream movement of goods between, for exam-
ple, suppliers and manufacturers. The related research focusing on
business-to-business (B2B) exchange, however, does not account
for either a new social dimension or the uncertainty associated
with crowdsourcing individual delivery agents (Ta et al. 2015).
Unfortunately, sourcing from the sharing economy for distri-
bution introduces additional risk (Ndubisi et al. 2016). While
CSL can facilitate collaboration among a retailer, independent
delivery agents, and the consumer, it also introduces competitive
consumption. Firms seeking delivery agents compete not only
with each other, but with driversother interests and needs. CSL
also introduces vehicle supply uncertainty not found in a pri-
vately owned eet because drivers manage their own schedules
and work as long or as little as they desire. As a result, the deci-
sion to use CSL involves navigating a trade-off between cost and
uncertainty. Furthermore, the comparison between CSL and a
dedicated eet is complicated by environmental factors, such as
demand and time windows, which can moderate the association
between vehicle supply uncertainty and logistics performance.
Operations researchers have explored the trade-offs between
owning and outsourcing transportation as part of the eet mix
problem (Hoff et al. 2010). These problems seek the most ef-
cient combination of nite resources, such as vehicles or techni-
cians, required to serve a customer population. A literature
review reveals several variants on the basic scenario, including
eet mix problems with demand variation (Topaloglu and Powell
8 V. E. Castillo et al.

To continue reading

Request your trial

VLEX uses login cookies to provide you with a better browsing experience. If you click on 'Accept' or continue browsing this site we consider that you accept our cookie policy. ACCEPT