Transportation Research Part D 26 (2014) 5259

Contents lists available at ScienceDirect

Transportation Research Part D

journal homepage: www.elsevier.com/locate/trd

A comparison of shipping containers from technical, economic

and environmental perspectives
Riccardo Accorsi, Riccardo Manzini , Emilio Ferrari
Alma Mater Studiorum, Bologna University, Italy

a r t i c l e

i n f o

Keywords:
Maritime transportation
Freight container
Temperature sensitive products
Food supply chain

a b s t r a c t
This paper compares standard, reefer, and thermal liner containers for the long-range
transportation of temperature-sensitive products. The thermal liner container is an alternative solution for maritime, rail and truck transportation between the dry and reefer containers. This study introduces a multi-criteria methodology to assess and compare
alternative container solutions that involves technical, economical, and environmental perspectives. The proposed methodology is applied to four signicant instances of maritime
shipments, which highlight the performances of alternative container solutions in reducing
the transport temperature uctuations. Economic aspects and transport environmental
impacts are also analyzed.
2013 Published by Elsevier Ltd.

1. Introduction
In the food sector, the increasing public awareness of the benets of healthier food has pushed the demand for agriculture
products for year-around availability. Thus, global food supply chains are expanding to match seasonal food production and
demand. The quality of such products is certied at the vendor dock (the production facility) but can be signicantly affected
by its transportation and other logistics activities. The consumer is seldom informed about the efciency, safety, and sustainability of the process that brings the product from the vendors to the place of consumption (Manzini and Accorsi, 2013).
Despite their origins, food products travel globally, and may experience diverse climates, weather conditions, handling
activities and logistics processes, including packaging, storage and distribution. The quality of these products depends on
the manufacturing activities as well as the packaging, handling and transportation. Specically, the role of packaging systems in logistics is central in the efcient management of any supply chains, and especially in the protection of product quality and shelf life, the compliance with standards and general rules, as well as concerns about environmental sustainability.
This study assesses the performances of freight container systems for long-range shipments. The increasing trend in global food distribution creates new challenges for shippers, importers, and logistics providers. Logistic managers and practitioners focus on choosing the proper distribution system, the transportation means, and the use of freight containers.
This study highlights the effects of transportation processes on temperature-sensitive products that are not recognized as
perishables by law. These products are usually shipped with reefer or controlled atmosphere containers, but sometimes
in dry container equipped with a wide variety of insulating materials. The choice of the system depends on the limitations
of the legal classication of products, which in turn depend largely on the water content of the product and its resulting
interaction with the environment humidity and temperature. For example, regarding high-water content products, the
Corresponding author. Tel.: +39 0512093406; fax: +39 0512093411.
E-mail address: riccardo.manzini@unibo.it (R. Manzini).

1361-9209/$ - see front matter 2013 Published by Elsevier Ltd.

http://dx.doi.org/10.1016/j.trd.2013.10.009

R. Accorsi et al. / Transportation Research Part D 26 (2014) 5259

53

cooling process is designed to inhibit the growth of decay-producing microorganisms and restrict enzymatic and respiratory
activities during the postharvest period1.
Given a particular product category, numerous alternative container solutions can be used to transport products, and
these can have widely different energy and related greenhouse gas (GHG) emissions proles. We use a multi-criteria perspective to choose the container system for long-range shipments that merges the concerns of the consumer (i.e., the quality
of products), the importer (i.e., the prot), and society (i.e., the impacts on environment).
2. Container systems
Global transportation is based on the assumption that an undamaged, tightly closed container protects its contents from
external inuences, such as ice, snow, rain, salt spray, dust, thermal radiation, ultraviolet light and other conceivable environmental inuences (Wild, 2012). However, many climatic and environmental stresses inuence transportation, such as
extreme temperatures or moisture, which can affect products and their packages. We consider the implications on the environment of three container systems.
 The standard container (SC), also referred to as a general-purpose container or dry container is suitable for the loading of
general cargo, with the majority of cargo being shipped with 20-foot (TEU) and 40-foot equivalent units (FEU) containers.
Standard containers protect goods from mechanical stresses and rainfall, reduce the inuences of moisture, and enable
handling by cranes.
 There are two main categories of reefer containers: porthole, which are not equipped with their own reefer systems but
supplied with cold air, and integrated systems, which are equipped with a refrigeration system powered by an external
source. TEU porthole containers consume an average of 12 kW, while integrated containers consume 36 kW. Over the
last decades, integrated containers (RC), which we focus on here, have become increasingly popular (Wild, 2012), being
widely adopted for the transportation of bananas, meat, citrus fruit, sh, and seasonal fruit.
 Thermal liner containers (TLC) are basic dry containers equipped with a thermal liner that can partially or completely
insulate cargo from climate stresses. We focus on thermal liners based on multi-sheet heterogeneous lms that consist
of two aluminum foils selected for their lightness, ductility, strength, resistance to environmental stresses, and ability to
protect from thermal shocks. The inner side of the liner consists of two polyethylene (PE) foils that, because PE is inert at
environmental temperatures, does not encourage the growth of algae or bacteria. PE also has low thermal conductivity
and high exibility enabling it to absorb mechanical stresses during handling and transportation activities. A woven fabric
core makes the liner sweat and avoids condensation enabling air circulation between the inner and outer sides of the
liner.
3. Methodology
The indoor temperature proles in maritime long-range shipments were tracked to measure the technical performance of
the system. The tracking is based on ad hoc thermometers embedded on the products carton, which trace the shipment from
the vendor facility to the importer docks. The sensor is a self-sufcient system that measures temperature and records the
result in a protected memory section. A mix of on-eld data and prior information is used to capture the economic performance associated with the shipping operations. The environmental performance results by a life cycle assessment (LCA)
methodology that estimates the carbon footprint of the alternative container systems during one shipment from a vendor
to the importer dock (Fig. 1).
The tracked shipments were selected based on the ports of origin and destination, the season of the year, and expected
temperature stresses (e.g., shipments across the equatorial belt). Two sensors2 per container recorded the temperature each
hour between 40 C and 85 C and were placed at the middle and the top of the pallet. Tracked proles are the average of the
data from the two sensors. The sensors were collected at the destination port. The tracking specically involved dry and insulated containers, taking the reefer container as the benchmark; it is assumed that the reefer container (RC) maintains a set temperature for duration of the shipment. Both containers were stowed on the upper deck and stacked in the middle to reduce the
inuence of direct radiation and to ensure well-ventilated conditions. The temperature proles are for the assessment of alternative container systems from technical, economic, and environmental perspectives.
4. Tracking
The tracking involved four shipments of bottled olive oil and wine, which departed from Italian ports for North America
and the Far East. Table 1 shows the travel characteristics.
1
For some examples of the link between the choice of containers, their handling and transportation and the quality and shelf life of food products, for edible
oils see Tsimis and Karakasides (2002) and Valli et al. (2013), for wines see Robinson et al. (2010), and for sh Margeirsson et al. (2012).
2
Each sensor integrates a 1-wire transmitter/receiver, a globally unique address, a thermometer, a clock/calendar, a thermal history log, and 512 bytes of
additional memory to store user data, such as a shipping certicate.

54

R. Accorsi et al. / Transportation Research Part D 26 (2014) 5259

Sensors
TEU SC

(2)

TEU TLE

Vendor

Importer
Shipment

(1)

Sensors

(3)

Laboratory

(4)

Fig. 1. Methodology.

Table 1
Tracking.
Shipment

Origin

Destination

Departing
date

Arrival date

Season

Max. temp.
(C)

Min. temp.
(C)

Ship. 1

Livorno (Italy)

28 July 2011

53.5

16.5

880

La Spezia
(Italy)
Livorno (Italy)

2 September
2011
12 September
2011
1 March 2012

Summer

Ship. 2

Detroit (via New York)

(US)
Tokyo (Japan) (via
Singapore)
Quebec (Canada)

Summer

46.0

21.5

1180

Winter

19.0

11.5

14 March 2012

Winter

40.0

3.0

Ship. 3
Ship. 4

La Spezia
(Italy)

Tokyo (Japan)

25 July 2011
30 January
2012
17 January
2012

Samples

730
1362

Fig. 2 illustrates the thermal proles that resulted from the tracking for SC and the TLC. The thermal prole is split into the
various logistics phases. Shipment 4 is reported without maritime transport, which is not useful for the comparative analysis.
Regarding SC, the shipments experience thermal shocks as extreme cold (Shipment 3), extreme heat (Shipment 1 and 2), or
deep thermal range (Shipment 4), all of which can affect the quality and the shelf life of bottled products, such as wine (Robinson et al., 2010) or olive oil (Valli et al., 2013). TLC protects such cargo from thermal uctuations.

5. Technical and economic perspectives

The assessment of the technical performance of alternative container systems is based on a key performance indicator
(KPI), with the thermal protection ability (TPA) being the ability of a specic container system to protect cargo from critical
temperature uctuations. Given a shipment and a container system, the evaluation of the TPA identies the quality of the
transportation process and the ability to protect the cargo from the vendor dock to the importer dock. The evaluation of
the technical KPI is based on the denition of the thermal benchmark, the goal temperature, which is the temperature at
the vendor facility. This temperature is tracked at the end of the manufacturing or packaging process and is assumed the
optimal temperature for the storage of products. Thermal uctuations are thus measured against the thermal benchmark
during transportation. Given a generic shipment, s, and a thermal range, DT, the TPA index is dened as:

Pns
TPAs;c;DT

i1

:j Txs;c;i j6 j TBs j DT

Ns

%

where xs,c,i is unity for sample i of shipment s in container c and zero otherwise; ns is the number of samples of shipment s;
T(xs,c,i) is the temperature of sample i in container c of shipment s; TBs is the hermal benchmark of shipment s; and DT is the
thermal range
The thermal range is the range of temperature that comprises the temperature samples near the thermal benchmark. The
higher the thermal range, the greater is the number samples involved. The benchmark is set for each shipment with ve signicant values of the thermal range considered. Fig. 3 shows the thermal benchmark for each shipment and illustrates the
trend of the TPA index given signicant thermal ranges of 1 C, 2 C, 3 C, and 4 C.
The value of the TPA index depends on the inuence of thermal uctuations around the benchmark during the shipment.
Given a thermal range, the value of the TPA positively correlates with the duration of travel at this temperature range. The
gure indicates that the TLC represents a good trade-off between SC and RC. In accordance with the temperature proles in
Fig. 2, the analysis can be re-designed by focusing on the most critical transportation processes, container stacking/handling

R. Accorsi et al. / Transportation Research Part D 26 (2014) 5259

55

Fig. 2. Tracking.

at port. Table 2 reports the results of the assessment of the TPA index for the four case studies, considering the samples
tracked during maritime travel.
Fig. 3 shows that SC and TLC perform according to the shipment prole, traveling season, and potential route. The TLC
showed better performances for Shipments 1, 2 and 3, and could protect cargo well from extreme heat and cold stresses,
but was less efcient for shipments with wide thermal excursions; Shipment 4 from 3 C to 40 C. Table 2 highlights
the inuence of maritime travel on the TPA index assessment. Given a thermal range, the TLC performs better in protecting
cargo from thermal stresses during maritime travel for summer shipments, Shipments 1 and 2, than for winter transports,
Shipments 3 and 4.
The choice of the container system directly inuences transportation costs. Here economic comparisons are based on data
from Banca dItalia (2012) and interviews with a group of vendors, carriers, and importers involved in the ocean consolidated
container service. In addition, the costs of transportation are affected by factors related to route, customs clearance, import/
export sanctions, and cargo characteristics and value.
The RC consists of a three-phase refrigeration unit and was primarily adopted in the cold chain for the transportation of
perishable products; being the most expensive container system, RCs are limited to perishable products, while dry freight is
shipped using SC. We focus on products that are dry load for the packaging (e.g., glass or plastic bottles), but which remain
climate-sensitive and affected by thermal stresses (e.g., edible oils, wine, cosmetics). For these, the TLC system can offer technical and economic trade-offs between the RC and the dry container (see Table 3).
Most containers used for global shipping belong to the carrier companies and are rented to importers and logistic providers. Leasing costs for a RC are higher than for a SC, while the TLC system is considered a basic SC equipped with a thermal
liner. In Table 3, operating costs for the SC and RC are accounted for leasing costs and allocated to one shipment. The values
refer to a general cargo TEU shipment exported from Italy to various geographical areas in different seasons. The analysis
only reports differential cost drivers for the comparison; e.g. terminal handling charges or international ship and port facility
security. Table 4 gives two examples of long-range shipments of Italian liquors and wines to the US and Far East. Even though
costs are estimated via interviews with exporters and logistics providers, the TLC container represents a good solution for
long-range shipments of temperature-sensitive products, especially if the value of the cargo is relevant.

56

R. Accorsi et al. / Transportation Research Part D 26 (2014) 5259

Fig. 3. TPA index for four tracked shipments and the three container systems.

Table 2
TPA index assessment and analysis.

DT

TPA (%)
With maritime travelling

Without maritime travelling

0 C

1 C

2 C

3 C

4 C

0 C

1 C

2 C

3 C

4 C

Shipment 1
SC (%)
TLC (%)
RC (%)

10
15
100

31
84
100

64
100
100

78
100
100

86
100
100

3
14
100

20
77
100

37
100
100

52
100
100

66
100
100

Shipment 2
SC (%)
TLC (%)
RC (%)

4
9
100

15
36
100

30
46
100

68
73
100

82
98
100

5
3
100

14
28
100

23
30
100

51
45
100

68
96
100

Shipment 3
SC (%)
TLC (%)
RC (%)

4
5
100

20
44
100

30
61
100

30
61
100

45
100
100

1
5
100

2
59
100

3
86
100

3
86
100

12
100
100

Shipment 4
SC (%)
TLC (%)
RC (%)

1
3
100

9
12
100

18
23
100

18
23
100

37
39
100

5
12
100

24
57
100

49
64
100

49
64
100

65
100
100

6. Environmental perspective
The environmental performance of the TLC system is evaluated using a LCA methodology that assesses products and
processes over the entire life cycle from a cradle to grave perspective. The analysis involves dening the, goal and scope

57

R. Accorsi et al. / Transportation Research Part D 26 (2014) 5259

Table 3
Economic comparison of container systems.
US $/TEU Shipment
Leasing costs
SC

Operating costs
Spring/summer

Autumn/winter

Mediterranean
Middle East
Africa
US/Canada
South America Pacic Coast
South America Atlantic Coast
Sud Est Asiatico
Far East
Oceania

540
990
1420
1480
1550
940
460
300
1550

530
1070
1380
1550
1690
950
450
340
1525

Mediterranean
Middle East
Africa
US/Canada
South America Pacic Coast
South America Atlantic Coast
Sud Est Asiatico
Far East
Oceania

650
1200
1700
1910
2040
1200
655
400
1970

Mediterranean
Middle East
Africa
US/Canada
South America Pacic Coast
South America Atlantic Coast
Sud Est Asiatico
Far East
Oceania

1010
1950
2670
2880
3090
1800
870
750
2925

TLC

RC

Assembling costs
Labor (0.5 h/person)

10

Maintanance costs
Labor (0.2 h/person)
Spare parts (80 gr. TL)

4
4

n.a.
n.a.
n.a.
n.a.
n.a.
n.a.
n.a.
n.a.
n.a.

Table 4
Economic comparison of different container systems in real shipments.
Case 1

Case 2

Port of Or.
Port of Dest.
Departure Date
Arrival Date
Travelling Season
n. of Items
Type of Iyems
n. of cases
Cargo weight (kg)
Cargo value (US$)

La Spezia (Italy)
Hong Kong
24 July 2013
18 August 2013
Summer
9
Wines & Liquors
1385
16,727
38,000

La Spezia (Italy)
Los Angeles (US)
3 September 2013
2 October 2013
Summer/Autumn
9
Wines & Olive Oils
1496
18,344
44,000

Costs (US$/Shipment)
SC
TLC
RC

300
418
750

1500
1928
2880

Costs/Value (%)
SC
TLC
RC

0.8%
1.1%
2.0%

3.9%
5.1%
7.6%

denitions, life cycle inventory (LCI), life cycle impact assessment (LCIA) and interpretation of results, the goal being to compare the impacts of the TEU shipment using alternative container technologies. The LCI consists of the denition of the system boundaries and the selection of datasets. The manufacturing of the TEU, including the supply of raw materials, melting,
bending and welding activities, is not considered in for either SC or RC, but amortized along their life cycle making it negligible for the selected functional unit (FU), one shipment. Taking the SC as the environmental benchmark, we consider the
differential drivers that can be associated with the FU, such as the energy supplied to the RC and the thermal liner for the
TLC.

58

R. Accorsi et al. / Transportation Research Part D 26 (2014) 5259

Raw Material
Supply

Raw Material
Supply

Raw Material
Supply

TL
manufacturing

TEU
manufacturing

Reefer Unit
manufacturing

SC

TLC

RC

Use phase

Use phase

Use phase

End-of-life

End-of-life

End-of-life

System boundaries
Environmental
benchmark

Fig. 4. System boundaries and LCA results.

Table 5
Results of the LCA of the TLC system

0%
0%
5%

1%

GWP100 kg CO2 eq

Phase

Impact category
Unit

GWP100
kg CO2
eq

Raw material supply

Aluminium foil
PE foil
Woven fabric
Thermoforming, with calendering/
RER U
Total manufacturing
Assembling
Maintenance
Total use phase
Waste transportation, Lorry 20-28
tons
Landll/CH U
Incineration/CH U
Total End-of-life

21.4111
8.2575
68.4624
5.3122

Aluminium Foil

1%

PE Foil

6%

Manufacturing

Woven Fabric

19%

Use phase

Manufacturing
7%

Maintanance

End-of-life

Assembling

Waste Transportation
61%

Landfill/CH U
Incineration/CH U

Total Thermal Liner (TL) Life

Cycle

103.4432
1.0363
0.0531
1.0895
0.2518
1.1195
6.6922
8.0635
112.5961

The LCI focuses on the thermal liner and considers the raw materials, manufacturing, use phase (i.e., assembling and
maintenance) and nal disposal. The carbon footprint (CF) is adopted as a metric to assess the environmental impact of
the FU with the TLC and RC taking the SC as environmental benchmark. The Intergovernmental Panel on Climate Change
2007 GWP 100 year-period was selected for the impact assessment. Although the CF provides a limited view of the overall
environmental impacts, and other methods consider many impact categories (e.g., human health, resource preservation, and
ecosystem quality), the equivalent carbon dioxide (CO2eq) has been widely adopted (Wright et al., 2011).
Fig. 4 gives the system boundaries. The thermal liner required to ll the TEU was split into aluminum foil, PE foil and woven fabric. The LCI also included the expected quantity required in the maintenance phase if the liner is ripped during the
assembling activities (see Table 5).
The assessment of the environmental impacts of the use phase of the RC system is based on calculating the energy supplied during transportation and handling activities as proposed in Table 5. We adopt Fitzgerald et al. (2011) metric that estimates the emission of CO2eq per maritime transportation of standard reefer container at 0.017 kg/tkm; Table 6 uses this to
estimate the environmental performances of the alternative system containers for the shipment objects of the tracking
campaign.
The table highlights the dependency of the RC environmental impact on the maritime route and the duration of handling
activities at ports. Assuming the SC as an environmental benchmark, the TLC represents an environmental care solution in
comparison with the RC.

59

R. Accorsi et al. / Transportation Research Part D 26 (2014) 5259

Table 6
Comparison of environmental performance.
kg CO2eq/TEU Shipment
SC
Use phase
Shipment 1
(7597 km)
Shipment 2
(14849 km)
Shipment 3
(7499 km)
Shipment 4
(17225 km)

SC
TLC
RC
SC
TLC
RC
SC
TLC
RC
SC
TLC
RC

Manufacturing

TLC
Use phase

End-of-life

103.44

1.09

8.06

RC
Use phase

0
2583
0
103.44

1.09

8.06
5049

0
103.44

1.09

8.06
2550

0
103.44

1.09

8.06
5857

Total
0.00
112.59
2582.98
0.00
112.59
5048.66
0.00
112.59
2549.66
0.00
112.59
5856.50

Saving TLC-RC (%)

95.64

97.77

95.58

98.08

7. Conclusions
Thermal protection ability, dened as the ability of a specic container to protect the cargo from critical temperature uctuations, is used to assess the technical performance of containers. The analysis assessed the performances of a standard,
integrated reefer, and thermal liner containers for four shipments of bottled wines and edible oils that were tracked from
Italian ports to North America and the Far East. In addition, the use of thermal lined containers is examined from operative
and environmental perspectives.
The results support the adoption of the TLC system for long-range maritime shipments to protect the cargo from climate
stresses and represents a good trade-off between technical and economic features to protect the quality of temperature-sensitive products and a lower impact on the environment than reefer containers.
Acknowledgements
The authors would like to thank Giorgio Gori S.r.l. for its involvement and support in this study, and Dr. Nino Caponi, Riccardo Pazzaglia and Alexander Braun for their valuable inputs into the project.
References
Banca dItalia, 2012. Indagine campionaria sui trasporti internazionali dellItalia. Benchmark 2011. Rapporto Finale.
Fitzgerald, W.B., Howitt, O.J.A., Smith, I.J., Hume, A., 2011. Energy use of integral refrigerated containers in maritime transportation. Energy Policy 39, 1885
1896.
Manzini, R., Accorsi, R., 2013. The new conceptual framework for food supply chain assessement. Journal of Food Engineering 115 (2), 251263.
Margeirsson, B., Lauzon, H.L., Palsson, H., Popov, V., Gospavic, R., Jonsson, M., Sigurgsladottir, S., Arason, S., 2012. Temperature uctuations and quality
deterioration of chilled cod (Gadus morhua) llets packaged in different boxes stored on pallets under dynamic temperature conditions. International
Journal of Refrigeration 35, 187201.
Robinson, A., Mueller, M., Heymann, H., Ebeler, S., Boss, P., Solomon, P., Trengove, R., 2010. Effect of simulated shipping conditions on sensory attributes and
volatile composition of commercial white and red wines. American Journal of Enology and Viticulture 61, 337347.
Tsimis, D.A., Karakasides, N.G., 2002. How the choice of container affects olive oil quality a review. Packaging Technology and Science 15, 147154.
Valli, E., Manzini, R., Accorsi, R., Bortolini, M., Gamberi, M., Bendini, A., Lercker, G., Toschi, T.G., 2013. Quality at destination: simulating shipment of three
bottled edible oils from Italy to Taiwan. La Rivista Italiana delle Sostanze Grasse XC (4), 163169.
Wild, Y., 2012. Container Handbook. Gesamtverband der Deutschen Versicherungswirtschaft e. V., German Insurance Association.
Wright, L., Kemp, S., Williams, I., 2011. Carbon footprinting: towards a universally accepted denition. Carbon Management 2, 6172.