ArticlePDF Available

Cost-effectiveness in the e-grocery business

  • RELEX Solutions


In the grocery business online start-ups, e-grocers, have in the last few years developed new business models to challenge the traditional bricks-and-mortar supermarkets. However, many of the big players in the e-grocery business, such as Peapod and Webvan in the USA, still operate in the red. They have financed rapid growth by collecting hundreds of millions of dollars from investors. These huge investments have enabled the e-grocers to enter the grocery market but, on the other hand, ineffective operations have used up capital on operating expenses. If the e-grocers want to become serious challengers to the traditional bricks-and-mortar business, they need to cut operational costs dramatically and fast. This paper illustrates some basic principles for cutting operational costs in the e-grocery business. The focus is on picking efficiency and order assembly costs. Solutions for achieving better picking efficiency are discussed. In addition, one detailed example of how a grocery distribution centre for efficient picking could be designed is presented.
Cost-effectiveness in
the e-grocery business
Vesa Ka¨ma¨ ra¨inen
Johanna Sma
Tomi Jaakola and
Jan Holmstro¨ m
Home delivery as such is not a new service
model in the grocery business. However, it
did not get widely spread until the Internet
became popular in the mid-1990s. At that
time, when the first e-grocers started to
operate, they offered Internet as a new tool for
ordering. That is, the ordering channel for the
groceries was new, but the operational
activities of home delivery were not.
From the customer’s point of view, home
delivery is a new service that makes life easier,
especially for people who have difficulties in
getting to a store or who do not have time for
shopping (see, for example, Morganosky and
Cude, 2000). Nowadays, e-grocers can also
offer other services, in addition to home
delivery. For example, it is possible to order
videos, dry cleaning or film processing in
addition to groceries (Ransdell, 1998).
Although e-grocery shopping has a lot to
offer, a problem is that consumer prices are
often higher than in traditional supermarkets.
This prevents fast growth and the creation of
an attractive and commercially profitable new
mass market. These high prices are a result of
inefficient home delivery and picking.
Some players, such as Streamline in the
USA and SOK in Finland, have developed a
solution for cutting delivery costs. The
groceries are delivered to locked reception
boxes, which are located in the customer’s
garage or yard (Streamline, 2000; S-Kanava,
2000). This means that the customer
becomes independent of the delivery
timetable, i.e. that the custom er does not
need to be at home to receive the goods. This
is also the most cost-effective home delivery
solution for the e-grocer. Results of an
analysis of home delivery in a suburban area
show a cost benefit of over 40 percent when
using reception boxes (Ka¨ma¨ra¨ inen et al.,
2000). The drawback is that this mode of
operation requires significant investment in
the reception boxes.
Tesco in the UK has attained near-national
coverage at a reasonable cost. It has chosen a
store-based fulfilment model that makes it
possible for the com pany to offer a full
product range and to use its existing assets to
the maximum (CIES, 2000). Tesco has a first
mover advantage in the UK and its model
allows it to quickly move into new areas and
cities. Thus, this is a fast way to increase the
sales volume. However, because products are
The authors
Vesa Ka
Èinen and Johanna Sma
Êros are
Researchers and Jan Holmstro
Èmis a Senior Research
Fellow in the Department of Industrial Engineering and
Management, Helsinki University of Technology, Finland.
Tomi Jaakola is a Logistics Designer at S-Kanava Oy,
Helsinki, Finland.
Grocery, Electronic commerce , Distribution centres,
Order picking
In the grocery business online start-ups, e-grocers, have in
the last few years developed new business models to
challenge the traditional bricks-and-mortar supermarkets.
However, many of the big players in the e-grocery business,
such as Peapod and Webvan in the USA, still operate in the
red. They have financed rapid growth by collecting
hundreds of millions of dollars from investors. These huge
investments have enabled the e-grocers to enter the
grocery market but, on the other hand, ineffective
operations have used up capital on operating expenses. If
the e-grocers want to become serious challengers to the
traditional bricks-and-morta r business, they need to cut
operational costs dramatically and fast. This paper
illustrates some basic principles for cutting operational
costs in the e-grocery business. The focus is on picking
efficiency and order assembly costs. Solutions for achieving
better picking efficiency are discussed. In addition, one
detailed example of how a grocery distribution centre for
efficient picking could be designed is presented.
Electronic access
The research register for this journal is available at
The current issue and full text archive of this journal is
available at
International Journal of Retail & Distribution Management
Volume 29 .Number 1 .2001 .pp. 41±48
#MCB University Press .ISSN 0959-0552
picked from existing supermarkets, operating
costs tend to increase faster than revenues at
high volumes. The issue is picking efficiency.
Webvan in the USA has improved picking
efficiency by investing in highly automated
distribution centres (Cuglielmo, 2000). Many
other e-grocery companies as well have
abandoned the early operating model of
picking from supermarkets and switched to
picking from special warehouses, so-called
distribution centres. The reason for this is
that picking is expensive when done in
conventional stores – custom ers are in the way
and the store layout is designed for displaying
products, not for picking efficiency. Existing
conventional stores’ structures have to be
significantly altered for attaining the efficient
picking and delivery mechanism that e-
grocery requires.
The distribution centre model requires
substantial investments, but also offers
opportunities to radically improve picking
efficiency and space utilisation as well as to
reduce loss due to pilferage or spoilage
(Ka¨ma¨ra¨inen, 2000). However, despite this,
most e-grocers still have trouble reducing
picking costs in practice. The problem is low
capacity utilisation, which means that the
overall operational efficiency remains poor.
This, in turn, means that the e-grocers cannot
provide competitive prices for the customers.
Furthermore, the service level offered by e-
grocers cannot improve radically before
resolving the operational issue of efficient
order picking. One of the e-grocer’s main
challenges is, therefore, improving overall
picking efficiency.
Cost-effectiveness through increased
picking efficiency
Order picking is one of the biggest cost drivers
in the e-grocery business. In bricks-and-
mortar supermarkets, the customers do their
own picking free of charge. The e-grocer,
however, has to pick the items for their
customers. This means that one of the biggest
e-grocery challenges is to increase picking
speed and, in this way, reduce the labour
costs of picking. Better picking speed can be
achieved by paying attention to who does the
picking as well as to the level of automation
(Reda, 1998).
First, who does the picking? There are two
alternatives: the existing supermarket or a
new specialist, the distribution centre. Using a
distribution centre rather than picking the
orders from conventional stores or wholesale
stores is potentially much more efficient.
Table I shows a comparison between a typical
supermarket and the e-grocery company
Streamline’s financial figures (assuming high
capacity utilisation for Streamline). The
distribution centre is essential for the
potential advantages of Streamline’s
operational model. Similar figures have also
been presented by, for example, Jaakola (1999).
Replacing stores with distribution centres
is potentially more efficient for several
reasons. First, picking the orders in
distribution centres rather than in stores
increases picking speed. This is because the
distribution centre can be especially design ed
for picking efficiency, whereas the
supermarket’s main concern is displaying the
products in an appealing way. In addition,
consumers doing their shopping do not get in
the way when the picking is done in a
distribution centre. Second, the labour costs
of a distribution centre are likely to be lower
than those of a supermarket. Finally, the
distribution centre allows for more efficient
space u tilisation (Jaakola, 1999). There are,
however, significant differences in cost-
effectiveness between different distribu tion
The second key factor to consider is the
level of automation. The level of automation
is important because by improving picking
speed it is possible to attain savings in
operating costs such as the salaries of the
picking personnel. At the moment, in e-
grocery companies the picking speed varies
between 100 and 450 lines per hour per
worker. The 100 lines per hour are picked
Table I Comparison of Streamline and a typical
Cost of goods sold 75 72
Operating costs 17 13
Distribution 4 6
Corporate overheads 3 3
Net profit (%) 1 6
Note: * Figures compiled by Smart Store, a research-
and-development initiative at Andersen Consulting
Source: Macht (1996)
Cost-effectiveness in the e-grocery business
Vesa KaÈ maÈ raÈ inen, Johanna SmaÊ ros, Tomi Jaakola and Jan HolmstroÈ m
International Journal of Retail & Distribution Management
Volume 29 .Number 1 .2001 .41±48
from a conventional store without any
automation (Ka¨ ma¨ ra¨inen, 2000), and the 450
lines per hour is a benchmark figure from
Webvan, which uses highly automated
distribution centres (Anders, 1999).
However, a high level of automation also
requires significant investments. For example,
Webvan’s distribution centres cost from $25
million to $35 million each, while less
automated distribution centres cost between
$ 4 million and $6 million (Cuglielmo, 2000).
At a capacity of 8,000 orders per day, seven
days per week, with an average order size of
about $103, Webvan could achieve an
operating margin of 12 per cent compared to
a 4 per cent margin for a traditional
supermarket. In reality, however, Webvan’s
distribution centre operated at a loss with a
less than 20 per cent capacity load, i.e. less
than 1,600 orders per day, five days a week in
the end of 1999 (Cuglielmo, 2000). The
example illustrates that efficient order picking
alone is not enough to make the e-grocery
business model cost-effective.
The main benefit of automation is the
reduced labour requirements, and thus
reduced labour costs. Next, we will
demonstrate how the maximum justifiable
investment in improving picking efficien cy in
different situations can be calculated. As an
example, we will look at a situation where we
have a distribution centre with the current
picking speed of 200 lines per worker per
hour. We would like to increase the picking
speed and are considering an investment to
increase the speed by 250 lines to 450 lines
per worker per hour. The employment cost is
$20 per hour. The expected lifespan of the
investment is eight years and the applied
discount rate 15 per cent.
In order for the investment to be profitable,
or at least not unprofitable, the savings it
causes have to at least equal the investment.
This means that the break-even point can be
expressed as follows:
Itot ˆStot;1
Itot ˆthe maximum total investment;
Stot ˆthe total savings during the lifespan
of the investment.
The annual savings, i.e. the annual reduction
in labour costs can be expressed as follows:
Sannual ˆT£p£ …W1¡W2;2
where T ˆthe amount of working hours
per worker per year,p
per hour,
W1ˆthe number of workers needed
without the investment,
W2ˆthe number of workers needed
after the investment.
The required number of workers can be
calculated by comparing the total picking
capacity needed to the picking capacity of
each individual worker:
where Ctot ˆthe total picking capacity
needed (lines per year),
Cind ˆthe picking capacity of each
worker (lines per year per
xˆthe picking speed (lines per
hour per worker).
The total capacity needed depends on the
amount of lines that need to be picked as well
as the capacity utilisation rate:
Ctot ˆL
where L ˆthe amount of lines that need to
be picked (lines per year);
Uˆthe capacity utilisation rate.
The net present value of the annual savings
during the lifespan of the investment can be
expressed as follows:
stot ˆX
where r ˆthe discount rate,
nˆthe expected lifespan of the
investment (years).
By combining formulae 1-5, we get the
following expression for the maximum total
investment that can be justified:
Itot ˆStot ˆX
³ ´:
In this case, if we make the simplifying
assumption that the capacity utilisation rate is
Cost-effectiveness in the e-grocery business
Vesa KaÈ maÈ raÈ inen, Johanna SmaÊ ros, Tomi Jaakola and Jan HolmstroÈ m
International Journal of Retail & Distribution Management
Volume 29 .Number 1 .2001 .41±48
1, the cost savings per line to be picked, i.e.
maximum investment per line, is therefore:
20 £450¡200
¡ ¢
The result shows that the maximum
investment in automation is $0.25 per line to
be picked – provided that capacity utilisation
is 100 per cent both before the investment
and after it. However, a 100 per cent capacity
utilisation rate is not realistic. Capacity
utilisation varies a lot depending on how
demand fluctuates over the week and during
the day. For example, in Finland about 40 per
cent of the grocery demand is concentrated to
Fridays and Saturdays (LTT, 1995). This
means that, if our capacity is inflexible, i.e. if
we use a lot of machines and automation, we
have to set the average capacity level
according to the peaks in demand, which
means that capacity utilisation is poor in the
beginning of the week. On the other hand, if
our capacity is flexible, i.e. if we use mainly
manual labour that only works and is paid
when it is needed, we can lower the average
capacity level needed (see Figure 1).
The capacity utilisation rate is, thus,
different in different situations and has to be
taken into account when considering the
profitability of an investment. When
calculating the maximum justifiable
investment, the same logic as presented above
can be used. However, the situation before
the investment is given a different capacity
utilisation rate than the situation after it. This
means that formula (6) changes as follows:
³ ´;7
where U1ˆthe capacity utilisation rate
before the investment,
U2ˆthe capacity utilisation rate after
the investment.
Figure 2 illustrates how the difference in
utilisation rates affects the value of the
investment. In the illustration, the capacity
utilisation rate before the investment is
assumed to be 1 – a simplification that makes
it easier to illustrate the changes. The original
picking speed is assum ed to be 200 lines per
As we can see, capacity utilisation has a
significant effect on cost savings. When
capacity utilisation increases, cost savings also
increase rapidly. To respond to the demand in
Figure 1 with same-day deliveries, a realistic
utilisation rate relying on a highly automated
solution is only 50 per cent (in a more manual
solution, workers can more easily be added to
meet peak demand than in a highly
automated one). In our example, the original
picking speed of 200 lines per hour needs to
be doubled to reduce the cost per order line at
a utilisation rate of 50 per cent. However,
with a utilisation of 60 per cent, a doubling of
the picking speed already yields a cost saving
in labour of $0.07 per line, and at utilisation
of 70 per cent the cost saving is $0.13 per
order line.
To summarise, automation is a solution for
achieving better picking efficiency in the e-
grocery business. However, if the e-grocery
Figure 1 Average weekly fluctuation in store visits in Finland in 1994
Cost-effectiveness in the e-grocery business
Vesa KaÈ maÈ raÈ inen, Johanna SmaÊ ros, Tomi Jaakola and Jan HolmstroÈ m
International Journal of Retail & Distribution Management
Volume 29 .Number 1 .2001 .41±48
companies fail to consider automation from a
capacity utilisation point of view, projected
cost savings will not be realised. Rather than
simply investing in automation, the e-grocers
also need to pay attention to the design
principles of the distribution centre. Special
attention must be given to the planning of the
order assembly process in the distribution
Design principles for better picking
The distribution centre should be seen as a
factory that assembles grocery orders rather
than a warehouse for storing goods. This
means that production principles
(Schonberger, 1982) such as just-in-time
(JIT) and lean manufacturing can be applied
in distribution centre design (Harmon, 1993).
Minimising handling and waste are powerful
concepts when making order picking m ore
When planning an e-grocery distribution
centre, the main goal should be to create an
efficient flow of products through the centre.
To obtain efficiency and not, for example,
allow slower-moving products to be in the
way when handling faster-moving products,
differences in product group characteristics
have to be considered when designing the
product flows and the layout of the
distribution centre.
The most important factor to consider is
the frequency of occurrence of different
products. This information can be obtained
by analysing consumer orders and
establishing in how many of the orders a
certain stock-keeping unit (SKU) appears.
Frequency of occurrence is more important
than volume because it directly affects the
amount of picking that has to be done – if
every order contains a certain item, it causes
more work than if every tenth order were to
contain ten pieces of this same item.
Êros et al. (2000) have presented a
model for consumer demand segmentation
(Figure 3) in which three demand segments
are used:
(1) continuous demand;
(2) occasional demand; and
(3) single purchase type of demand.
The model and its three segments can also be
used to describe how products, or SKUs, with
different frequencies of occurrence can be
grouped. Items with high frequency and
Figure 2 Total cost of savings per line, i.e. the value of the investment, at different rates of capacity utilisation
Figure 3 Demand segmentation of products
Cost-effectiveness in the e-grocery business
Vesa KaÈ maÈ raÈ inen, Johanna SmaÊ ros, Tomi Jaakola and Jan HolmstroÈ m
International Journal of Retail & Distribution Management
Volume 29 .Number 1 .2001 .41±48
relatively stable demand belong to the
continuous demand group. Items that have a
lower frequency of occurrence belong to the
occasional demand group. Finally, items that
only rarely appear in the consumers’ orders
belong to the group called single purchase.
The layout of the distribution centre should
take these groups into consideration.
Handling and movement of the continuous
demand products must be minimised. Ideally,
the products are cross-docked in the
distribution centre. There is no need to store
these items on the shelf. Products that belong
to this group must be placed near to inbound
and outbound areas. In Figure 4, a layout
solution for these items is presented.
Products that belong to the occasional
demand group are unpacked from pallets
(wholesale packages) to smaller units. This
makes it possible to save space. Items are
stored in flow racks and on open shelves. U-
shaped picking cells (Schonberger, 1982) of
compatible products can be formed to
increase picking efficiency (Figure 5). To
avoid crossing paths, the shelves are loaded
and unloaded from different sides. The aim is
to get these items to flow through the
distribution centre.
There is no need for forming cells for
products that demonstrate single purchase
type of demand, due to their rare occurrence
in the orders. The most important thing as far
as this group is concerned is to minimise the
risk of stock-outs – it is important that the
customers’ orders can be filled completely in
order to attain good service quality (Sakki,
1999). In Figure 6, the solution for storing
items that have low frequency of occurrence is
Although frequency of occurrence forms
the basis for the product flow and layout
design, there are also other factors, such as
the preservation temperatures and physical
characteristics of the products to consider.
When dealing with groceries, preservation
temperatures are crucial. In the distribution
centre, proper preservation of the goods
requires that there should be at least three
different temperatures in use:
(1) room temperature;
(2) chilled; and
(3) frozen.
This increases the complexity of grouping and
layout planning. For instance, items that
belong to the continuous demand group have
to be placed in at least three different places
because of preservation temperatures. The
temperature aspect also affects what shelf and
technology solutions can be used. For
example, frozen items may need to be placed
so that the picker does not have to do the
picking inside a freezer. Picking should be
possible to do outside the freezer, through
special Plexiglas doors.
Furthermore, the physical characteristics of
the products have to be taken into account.
Some items, such as meat and detergents,
may not be stored in the same place. The
weights, dimensions and durability of the
products also affect layout planning. By
placing items that have similar properties in
the same area, the picking routes can be
planned so that heavy items are picked first
and fragile items will be situated on top of the
picking tote.
Picking the products is, however, just one of
the distribution centre’s functions. The
products also have to be assembled into
Figure 4 Solution for continuous demand products
Cost-effectiveness in the e-grocery business
Vesa KaÈ maÈ raÈ inen, Johanna SmaÊ ros, Tomi Jaakola and Jan HolmstroÈ m
International Journal of Retail & Distribution Management
Volume 29 .Number 1 .2001 .41±48
complete orders. Figure 7 presents how
assembly can be arranged efficiently. Items
are moved to the assembly area just before
delivery in order to maintain the appropriate
preservation temperature as long as possible.
By arranging picked order totes according to
customers’ locations, for example postal
codes, it is possible to simplify the final
assembly process (Harmon, 1993).
The distribution centre does not need any
extraordinary technology to be effective. The
most important point to understand when
improving picking efficiency is to realise that
shelves and other facilities do not need to be
automated. The first priority is not
automation, but organising the distribution
centre material flows and layout for efficient
order assembly. The proper reference point
for designing a distribution centre for home
delivery is not a wholesaler or grocery
manufacturer distribution centre, but an
assembly plant for make-to-order PCs or
Segmenting products according to the
frequency in the shopping baskets of the
consumer makes it possible to significantly
improve efficiency using only ordinary picking
shelves and gravitation flow racks. The next
level of improvement is achieved by using bar-
codes, pick-to-light systems and a warehouse
management system. With investments in
technologies that frees both hands of the
picker, it is possible to make the picking
process much more efficient, again without
investment in automatic handling systems.
E-grocers try to challenge the traditional
bricks-and-mortar grocery business, but their
operational costs are still too high. In addition
to home delivery, order picking is currently
Figure 7 Order assembly solution
Picking cells for occasional demand items
Figure 6 Solution for single purchase products
Cost-effectiveness in the e-grocery business
Vesa KaÈ maÈ raÈ inen, Johanna SmaÊ ros, Tomi Jaakola and Jan HolmstroÈ m
International Journal of Retail & Distribution Management
Volume 29 .Number 1 .2001 .41±48
far too inefficient and expensive. One of the
main e-grocery challenges of today is,
therefore, to increase picking speed.
Questions that need to be addressed are: who
performs the picking (existing supermarkets
or new specialists, distribution centres) and
what level of automation is desired?
In this paper, we have shown that, when
volumes are sufficient, distribution centres
become more efficient than supermarkets for
several reasons. One of the main benefits of
the distribution centres is the opportunity to
significantly increase the picking speed.
Automation is the obvious solution for
further increasing the picking speed. This,
however, requires significant investments. In
addition, as we have shown is this paper,
projected savings may not be realised if the
full capacity of the distribution centre is not
efficiently utilised. Unfortunately, due to
fluctuating demand, the capacity utilisation is
often rather poor in the grocery business. This
diminishes the attractiveness of investing in
Therefore creating flexible and expandable
product flow and layout designs should be
regarded as more important than rapidly
increasing the level of automation. The goal
should be a distribution centre through which
the products flow efficiently and without
stopping unnecessarily, and in which changes
and adjustments, such as changes in the
placements of the products or in the size of
the operation, can be made continuously as
the business develops. Instead of taking the
risk of tying a lot of money to automation that
we cannot fully exploit, we should create an
opportunity to take smaller steps as the
business grows.
Finally, before embarking on automating
order assembly, there is much to do for e-
grocers in developing the service model. The
point is that the potential utilisation rate for
automated solutions can be improved by
levelling the demand. Changing the service
model can level demand for grocery products
in several ways. For example, to level the
demand during the day a reception box in the
garage or front yard of the consumer is useful.
The e-grocer can level the load on the
distribution centre by delivering orders to the
reception box before the peak hours in the
afternoon when people return from work.
Also, replenishing automatically the
household with continuous demand products
can be effective to level demand during the
week. This way, changing the service model
to level demand can make investment in
improving picking speed more attractive . The
result is that service and cost-effectiveness can
be developed simultaneously.
Anders, G. (1999), ``Will Net grocery shopping ever take
CIES (2000), ``A perspective on the key issues in the
development of food home shopping in the UK’’,
presented by John Harvey at CIES Supply Chain for
E-commerce and Home Delivery in the Food Industry
Conference, Berlin, Germany, 18-19 May.
Cuglielmo, C. (2000), ``Can Webvan deliver the goods?’’,
Inter@ctive Week, 7 February, http://
Harmon, R.L. (1993), Reinventing the Warehouse, The
Free Press, New York, NY.
Jaakola, T. (1999), ``Verkkokaupan paikallinen
jakelukeskus’’, Master’s thesis, Tampere University
of Technology, 93 pp.
KaÈmaÈ raÈ inen, V. (2000), ``Internet muuttaa jakelurakenteita
± elektronisen paÈi vittaÈ istavarakaupan
liiketoimintamallit ja loppujakelupaika n valinta’’,
Master’s graduate thesis, Helsinki School of
Economics and Business Administration, 91 pp.
KaÈmaÈ raÈ inen, V., Saranen, J. and HolmstroÈ m, J. (2000),
``The reception box impact on home delivery
efficiency in the e-grocery business’’, Working
paper, Helsinki University of Technology.
LTT (1995), Kotitalouksien paÈ ivittaÈ istavaroiden ostotava t
1994 sekaÈ ostostenteon ongelmat,
Liiketaloustieteellinen tutkimuslaitos, LTT Sarja
B121, Helsinki.
Macht, J. (1996), ``Errand boy’’, Inc, Vol. 18 No. 16,
pp. 61-6.
Morganosky, M.A. and Cude B.J. (2000), ``Consumer
response to online grocery shopping’’, International
Journal of Retail & Distribution Management,
Vol. 28 No. 1, pp. 17-26.
Peapod (2000),
Ransdell, E. (1998), ``How do you win on the Web?’’, Fast
Company, August.
Reda (1998), ``Internet food retailers face tough picking,
delivery issues’’, Stores, March.
Sakki, J. (1999), Logistinen prosessi, tilaus-toimitusketju n
hallinta, Jouni Sakki Oy, Espoo.
Schonberger, R. (1982), Japanese Manufacturing
Techniques: Nine Hidden Lessons in Simplicity, The
Free Press, New York, NY.
S-Kanava (2000),
SmaÊros , J., HolmstroÈ m, J. and KaÈmaÈ raÈinen , V. (2000),
``New service opportunities in the e-grocery
business’’, International Journal of Logistics
Streamline (2000),
Webvan (2000),
Cost-effectiveness in the e-grocery business
Vesa KaÈ maÈ raÈ inen, Johanna SmaÊ ros, Tomi Jaakola and Jan HolmstroÈ m
International Journal of Retail & Distribution Management
Volume 29 .Number 1 .2001 .41±48
... Warehousing and order picking costs are very high in traditional but also in logistics systems for home delivery, in which they have the largest share in total costs together with delivery costs (Kämäräinen et al. 2001b;Punakivi & Tanskanen 2002;Kämäräinen & Punakivi, 2002;Vanelslander et al., 2013). The costs of picking a wide range of individual products in small quantities are higher than the picking costs at the pallet level (Agatz et al., 2008b;Vanelslander et al., 2013). ...
... Order picking can be realized in the sales or storage space of the store. Picking in sales space (taking off the shelf) is expensive and inefficient (Kämäräinen et al., 2001b), because it is not designed for this activity (Kämäräinen & Punakivi, 2002;De Koster & Neuteboom, 2001), and can disturb the functioning of stores and the standard of service for online customers, as well as for customers who make traditional purchases. Picking inefficiencies are caused primarily by inadequate locating of related goods (e.g. ...
... Automated or semi-automated picking is applied in dedicated centers (Hübner et al., 2016;Kämäräinen & Punakivi, 2002). The main advantages of automated picking are reduced labor needs and lower operating costs (Kämäräinen et al., 2001b). Another advantage of using a dedicated LC is the convenience of applying the VMI concept (subsection 2.1) (Kämäräinen & Punakivi, 2002). ...
Full-text available
Home delivery, as a flow of goods between the company and the consumer (B2C) at the end of the supply chain ("last mile"), can be realized by the manufacturer, retailer or third party (logistics provider, crowd-worker). The growth of home delivery services in recent decades has been predominantly stimulated by the development of online ordering, i.e. e-commerce. Delivery characteristics depend on the characteristics of the environment and a number of factors related to stakeholders (customers, companies, governments). In order to improve the quality of service and customer satisfaction, company profitability and (or) reduce the negative effects of deliveries on the environment, area and community functionality, various delivery models are designed and implemented in terms of need for ordering, frequency, ordering and payment system, starting and end points, executors, reception method, security, delivery area, speed and time of realization, return flows, etc. This paper proposes a framework for a comprehensive structuring and classification of home delivery models according to the stated characteristics/criteria. Also, the advantages, disadvantages, interdependence and applicability of different delivery models in changing circumstances are described. In this way, a comprehensive review of home delivery flows was performed, the literature that deals with this area in a similar way was supplemented, but the basis for future research was also created.
... A dedicated fulfilment centre is typically designed to achieve efficient picking operations while high-street stores are designed primarily for displaying and selling products to customers (Peters, 2000;Yrjola, 2001). Some of the picking operations may even be automated in the 56 Chapter Two: Literature Review dedicated fulfilment centre (Kamarainen et al., 2001b). However, investments on a dedicated fulfilment centre are higher than when operating from an existing store. ...
p>Home shopping and delivery services offer customers the opportunity to purchase goods and receive deliveries to their home rather than having to travel to high-street stores. Given the promising future of home shopping and delivery market, many efforts have been devoted to solving the problems currently encountered by service providers and customers which include unsecured deliveries, first-time delivery failures, demands for faster delivery, and product returns. Of major concern in this research are the implications of home delivery failures when there is nobody in to receive the package at the delivery address. Collection/delivery point (CDP) systems are one of the emerging solutions to mitigate failed home deliveries, in which CDPs are used as alternative addresses to receive the packages. Particularly focused on the small package home shopping market, this research has identified and modelled the existing home delivery and CDP methods. The carrier and customers travelling distance incurred in each delivery method was compared. It was then possible to quantify whether the CDP method is an economic solution to improve home delivery operations and the environment. A six-step research method was then developed to achieve those research objectives. Firstly, the existing and emerging home delivery methods were identified from the literature. The second stage consisted of conducting two home delivery surveys in Winchester and West Sussex, respectively. The surveys were'used to identify the home shopping and delivery characteristics of customers. In the third research step, the</p
... Due to this easiness then no customers lean on a particular e-grocer. Customers can visit any of the e-grocers to browse many of product catalogs, do price comparison, and purchase products [4]. Many grocery businesses became as a e-grocers and developed new business models to challenge conventional grocery retail. ...
Full-text available
As the number of conventional grocers transformed as e-grocers, which makes customers much more comfortable purchasing online groceries. However, e-loyalty always becomes a challenge for e-grocers; thus, it's a crucial thing to have understanding about e-loyalty of Indonesian's customers are formed. The main objective of this research investigates big data recommendation, e-tailing quality, and e-satisfaction that affect customer intentions to form e-loyalty of e-grocery business in Indonesia. A quantitative method as an empirical research was conducted using an online survey from major cities in Indonesia who have shopped at least twice at any e-grocers as analysis units were obtained. This research has a contribution to understanding the drivers of e-loyalty in the e-grocery business; also, e-grocer management can focus on the e-grocery platform by providing the best quality design, security, product availability and customer service.
... Regarding the warehouse preparation model, the existing studies have focused on improving picking efficiency, as it is a great advantage over store preparation. In this sense, Kämäräinen et al. (2001) have focused their research on optimising order preparation and consolidation processes. With the aim of reducing operational costs, these authors focus on the design of the picking area. ...
Full-text available
Traditional supermarket chains that are adopting an omni-channel approach must now carry out the order picking and delivery processes to serve online orders, previously done by the customer. The complexity of the logistics processes has increased, therefore modelling and optimising e-grocery operations becomes definitely important. Since there are few studies modelling order picking and delivery processes, we propose an approach that simultaneously optimises the decision variables of different functions which have traditionally been treated separately. In this study, we present a linear programming model for store-based e-fulfilment strategies with multiple picking locations. The proposed model optimises the allocation of online orders to stores, based on the e-fulfilment costs. As well as minimising the picking and delivery costs, the proposed approach consolidates workloads in order to avoid idle times and reduce the amount of resources required. A weighted sum method is applied to compute the solution, integrating parameters that represent different store features such as the product range, sales mode and physical store activities. The proposed model has been tested on one of the largest grocery sellers, showing that substantial savings can be achieved by reallocating orders to different stores, time windows and delivery vehicles. By focusing on optimising e-fulfilment resources, this approach serves as a guide for traditional grocery sellers to redesign their supply chains and to facilitate decision-making at a managerial level.
... In addition to attractive and easily navigable websites (Freeman & Freeman, 2011), one key to success in e-grocery shopping is low operational costs in order to offer competitive prices and effective delivery services (Anckar, Walden, & Jelassi, 2002;Kamarainen, Smaros, Jaakola, & Holmstrom, 2001). This lesson was learned by both traditional grocers and mega retailers such as Walmart or Target, and Amazon, which acquired Whole Foods in 2017 to boost its physical presence. ...
In spite of the popularity of e-shopping, only 16% of US adults have ordered groceries online, and 7 out of 10 of those who currently buy groceries online do so at most twice a month. Understanding the determinants of e-grocery shopping is important for grocers, supply chain managers, and urban planners. In this context, we first explore how deliveries from online shopping have been changing over time. From our analysis of the 2009 and 2017 National Household Travel Surveys, we found that online shopping has been embraced by increasingly diverse households, although income, education, and some racial/ethnic differences persist. Our analysis of the 2017 American Time Use Survey shows that Americans are 24 times more likely to shop for groceries in stores than online. Moreover, in-store grocery shoppers are more likely to be female and unemployed, but less likely to belong to younger generations, to have less than a college degree, or to be African American. The gender imbalance in grocery shopping is larger online than in stores, but e-grocery shoppers do not otherwise differ from the general population. Future travel and e-shopping surveys (especially for e-grocery) should combine time use and travel questions with retrospective questions about online purchases.
... To solve last-mile problems, researchers, institutions, e-commerce companies (ECs), logistics service providers (LSPs), community service providers (CSPs), and many other stakeholders have eagerly worked together for decades and come up with diverse innovations relating to delivery routing, vehicles, and reception alternatives [7,8]. Customer pick-up at collection and delivery point (CDP), as a form of unattended reception without face-to-face contact with couriers [9], has been a hotspot in logistics research and practice as a sustainable, propagable, and widespread solution, with generally accepted advantages in economic efficiency [10,11], environmental friendliness [12,13], social values [14,15], and service quality [16,17]. Thus far, typical types of CDP include automated parcel stations (APSs), specialized manned ones (SM-CDP) such as post offices, and unspecialized manned ones (USM-CDP) such as convenient stores, which are becoming key features of e-commerce's and logistics players' strategies [18] and an important component in the delivery networks of urban areas in many countries, such as France, Germany, the UK, and Japan [19,20]. ...
Full-text available
In recent years, customer pick-up at collection and delivery points has become a popular alternative to traditional home delivery, which is under great pressure. However, current service of pick-up facilities has seldom been geographically evaluated despite its general uneven distribution and diverse needs. In this paper, in order to interpret the differentiation in customers’ service demands toward reception alternatives and in facilities’ service excludability in different built environments, a two-step floating catchment area (2SFCA) method is improved to measure customers’ spatial accessibility to pick-up facilities, providing a methodology to evaluate the match relation between the differentiated supply and demand of pick-up service. A case study of widespread automated parcel stations (APSs) is conducted in Hangzhou, China and correlative factors to residents’ accessibility are discussed. From the results, residents’ accessibility to pick-up service shows significant spatial unevenness and social inequity in the study area, which is found to correlate most to residences’ maintenance management. As well-managed, gated communities generally hold effective access to exclusive services, most open communities and self-built, single houses are in need of improvement due to inadequate service stemming from a high aging rate, lack of property management, and low service availability of nonexclusive facilities in open areas.
Purpose This paper investigates the logistics management in the e-grocery sector. It contrasts the key issues faced by practitioners and the topics addressed in the academic literature, to identify potential misalignments between research and practice and propose avenues for future efforts. Design/methodology/approach This work adopts a twofold methodological approach. From an academic perspective, a systematic literature review (SLR) is performed to define the topics addressed so far by scholars when analysing e-grocery logistics. From a managerial perspective, a Delphi study is accomplished to identify the most significant issues faced by logistics practitioners in the e-grocery context and the associated significance. Findings The study develops a conceptual framework, identifying and mapping the 9 main logistics challenges for e-grocery along 4 clusters, in the light of a logistics-related revision of the SCOR model: distribution network design (area to be served, infrastructures), order fulfilment process (picking, order storage, consolidation, delivery), logistics-related choices from other domains (product range, stock-out management) and automation. These elements are discussed along three dimensions: criticalities, basic and advanced/automation-based solutions. Finally, the main gaps are identified – in terms of both under-investigated topics (order storage and stock-out management) and investigated topics needing further research (picking and automation) – and research questions and hypotheses are outlined. Originality/value This paper provides a threefold contribution, revolving around the developed framework. First, it investigates the state of the art about e-grocery logistics, classifying the addressed themes. Second, it explores the main issues e-grocery introduces for logistics practitioners. Third, it contrasts the two outcomes, identifying the misalignment between research and practice, and accordingly, proposing research directions.
Purpose This paper identifies, configures and analyses a solution aimed at increasing the efficiency of in-store picking for e-grocers and combining the traditional store-based option with a warehouse-based logic (creating a back area dedicated to the most required online items). Design/methodology/approach The adopted methodology is a multi-method approach combining analytical modelling and interviews with practitioners. Interviews were performed with managers, whose collaboration allowed the development and application of an empirically-grounded model, aimed to estimate the performances of the proposed picking solution in its different configurations. Various scenarios are modelled and different policies are evaluated. Findings The proposed solution entails time benefits compared to traditional store-based picking for three main reasons: lower travel time (due to the absence of offline customers), lower retrieval time (tied to the more efficient product allocation in the back) and lower time to manage stock-outs (since there are no missing items in the back). Considering the batching policies, order picking is always outperformed by batch and zone picking, as they allow for the reduction of the average travelled distance per order. Conversely, zone picking is more efficient than batch picking when demand volumes are high. Originality/value From an academic perspective, this work proposes a picking solution that combines the store-based and warehouse-based logics (traditionally seen as opposite/alternative choices). From a managerial perspective, it may support the definition of the picking process for traditional grocers that are offering – or aim to offer – e-commerce services to their customers.
Conference Paper
Full-text available
Customers are becoming increasingly demanding and logistics is gaining more importance in order to assure customer satisfaction in terms of money, quality and time. The goal of this paper is to show that warehouses can be a source of competitive advantage and that a critical approach towards the existing available space may lead to increases in capacity with low investment. To accomplish this, a company is used as a model and suggestions for improvements are carried out. Regarding the capacity, a volume approach is taken into consideration, both in terms of the products and regarding the bins. Moreover, opportunities to achieve better space usage are presented by suggesting additional or different bins. Therefore, it is possible to reduce costs, to increase safety, to increase capacity up to 9,77%, and to help minimizing traveling distances and damages caused to the items.
Purpose Grocery sellers that have entered the online business must now carry out order fulfilment activities previously done by the customer. Consequently, in a context of online sales growth, the purpose of this study is to identify and implement best practices in order to redesign the order picking process in a retailer with a store-based model. Design/methodology/approach To identify different work alternatives, an approach is developed to analyse the methods used in distinct stores of one large Spanish grocer. The methodology employed is a three-step statistical analysis that combines ANOVA and MANOVA techniques to settle on the best alternatives in each case. Findings Substantial improvements can be achieved by analysing the different working methods. The three-step statistical analysis identified best practices in terms of their impact on preparation time, allowing a faster working method. Practical implications To manage business processes efficiently, online grocers that operate store-based fulfilment strategies can redesign their working method using a criterion based on their own performance. Originality/value This is one of the few contributions focusing on the improvement of e-grocery fulfilment operations by disseminating best practices through decision-making criteria. This study contributes by addressing the lack of approaches studying the order picking process by considering its various features and applying best practices.
Full-text available
Home delivery of groceries is not yet a very popular service among consumers. One reason for the slow progress has been the time-consuming and expensive ordering process. In recent years, Internet-based solutions have solved most of the problems related to the order transaction process – making ordering simpler, cheaper and faster. However, there are still a lot of unsolved problems in the e-grocery business. One of the biggest obstacles is inefficient home delivery. This paper examines how different solutions for goods receipt affect home-delivery efficiency. Different alternatives for receiving the goods are presented and the service levels of these concepts are described from the consumer’s point of view. In addition, the costs for the e-grocer are studied. The efficiency of using a reception box is demonstrated by simulating two alternative receiving concepts.
Full-text available
Reports a preliminary assessment of consumer response to and demand for online food retail channels. Data were collected from 243 US consumers who currently buy their groceries online. The majority of online users were younger than 55 years of age, female, and reported annual incomes of $70,000 or more. Over 70 percent reported convenience and saving time as their primary reasons for buying groceries online but 15 percent cited physical or constraint issues that made it difficult for them to shop at grocery stores. Of the respondents, 19 percent bought all of their groceries online. Also reports demographic and online shopping variables that are significantly related to the primary reason for shopping online, willingness to buy all grocery items online, perception of time spent shopping online vs in the store, and experience with online grocery shopping.
Full-text available
Currently, efforts in the e-grocery business focus on improving the purchase transaction and physical distribution of goods. However, simply improving ordering and fulfillment does not make e-grocery shopping a viable competitor to the current supermarket business model. To become a profitable growth business, the e-grocers have to offer their customers more value. It is not enough to offer customers a range of physical products. A range of new meaningful services is also needed. This article investigates how such new, breakthrough services can be developed. A framework for systematically examining customer demand and identifying corresponding services is presented. The importance of the e-grocer being able to offer the right mix of services to meet the customers' individual and changing needs is demonstrated. Concrete examples of both new services and a service mix are described.
Full-text available
IN this article, Professor Schonberger summarizes his book of the same title (published by The Free Press in 1982). The emphasis is on the advantages of "just-in-time" production and total quality control, and what just these two techniques can do for manufacturing. Richard Schonberger is a professor in the Department of Management at the University of Nebraska--Lincoln. His research interests have lately been focused on the strengths and weaknesses of Japanese manufacturing techniques, particularly in comparison to American techniques.
Will Net grocery shopping ever take off?
  • G Anders
A perspective on the key issues in the development of food home shopping in the UK
  • CIES
Can Webvan deliver the goods?
  • C Cuglielmo
Cuglielmo, C. (2000),``Can Webvan deliver the goods?'', Inter@ctive Week, 7 February, http://
Reinventing the Warehouse
  • R L Harmon
Harmon, R.L. (1993), Reinventing the Warehouse, The Free Press, New York, NY.
Verkkokaupan paikallinen jakelukeskus
  • T Jaakola
Jaakola, T. (1999),``Verkkokaupan paikallinen jakelukeskus'', Master's thesis, Tampere University of Technology, 93 pp.
Internet muuttaa jakelurakenteita – elektronisen päivittäistavarakaupan liiketoimintamallit ja loppujakelupaikan valinta
  • V Kämäräinen