How New Technologies Can Improve Cold Chain

Management?

Luigi Battezzati, Giovanni Miragliotta, Alessandro Perego and Angela Tumino

Department of Management, Economics and Industrial Engineering

Politecnico di Milano, 20133 Milano – Italy

Abstract. There is a growing attention towards technological solutions that can

improve cold chain management. This paper analyzes pros and cons of four

alternative solutions (Data Loggers, Time Temperature Indicators, semi-passive

RFId and Wireless Sensor Networks), paying particular attention to the overall

balance of costs and benefits. Thanks to this preliminary study, a solution based

on Wireless Sensor Networks has been designed to monitor more efficiently the

ice-cream supply chain of a prominent company (Nestlé Italy) and its impacts

on overall cold chain performance have been evaluated.

1 Introduction

Cold chain management is gaining increasing attention both in the academic and

generalist press. There are several industries (e.g. pharmaceutical, healthcare, fresh

and frozen foods) in which an unbroken cold chain must be guaranteed, i.e. an

uninterrupted series of storage and distribution activities to be performed at a given

temperature range.

From a technological perspective, there are several solutions available that can help

monitor and control the temperature. In addition to the most traditional, there are

some innovative systems based on Radio Frequency Identification (RFId) technology

that enable new functionalities and opportunities. In the last few years the

potentialities of RFId technology have started to be explored [1], but today much of

the attention is still paid to applications that improve the efficiency (e.g. reducing the

identification time) and the accuracy (e.g. reducing the percentage of errors) in the

Consumer Packaged Goods supply chain (e.g. [2], [3]).

This study aims at analyzing how innovative technologies can be used to manage a

cold chain in a more efficient (e.g. by reducing the costs of temperature controls) and

effective (e.g. by assuring a better product quality) way. Among the above mentioned

industries, this paper focuses on the food industry, and in particular on the ice-cream

industry, even though almost every issue discussed is common to other fresh (and

frozen) cold chains whose products suffer frequent temperature abuses and shocks

[4]. In this regard, the purpose of this paper is twofold: first, it presents an analysis of

the pros and cons of four temperature monitoring technologies, from the most

traditional (e.g. Time Temperature Indicators and Data Loggers) to the most

Battezzati L., Miragliotta G., Perego A. and Tumino A. (2008).

How New Technologies Can Improve Cold Chain Management?.

In Proceedings of the 2nd International Workshop on RFID Technology - Concepts, Applications, Challenges, pages 149-156

DOI: 10.5220/0001735901490156

 SciTePress

innovative ones (e.g. semi-passive RFId and Wireless Sensor Networks, WSN).

Second, it seeks to analyze in depth how a solution based on Wireless Sensor

Networks can improve the ice-cream cold chain management of a prominent company

– i.e. Nestlé Italy.

In this regard, the paper is structured as follows: Section 2 describes the research

context while section 3 introduces the four technological solutions available, pointing

out their advantages and disadvantages as well as their requirements in terms of costs

and organizational complexity; Section 4 analyzes the most innovative WSN solution

whose application to the Nestlé Ice-Cream supply chain in Italy is then discussed in

Section 5. Eventually Section 6 is devoted to draw some concluding remarks and

outline future research paths.

2 Context and Problem Setting

Figure 1 reports the reference supply chain, which reflects the structure of the most

common ice-cream cold chain.

First‐tierwarehouses

Plants

Second‐tierwarehouses(dealers/distributors)RetailerDistributionCenters

Cafés/Restaurants/Themeparks

RetailerPointofSale

PoS 1)

PoS 1

PoS n

Special clients

RETAIL

IMPULSE

Fig. 1. The reference supply chain.

After the ice-creams have been produced and palletized, they have to spend a

certain amount of time – depending on the type of product – in the plant warehouse

for the hurdling phase, to ensure that the required (-18°C) is reached. After the

hurdling, the products are shipped to the first-tier warehouses which have two main

functions: to collect the products manufactured in different plants and to stock them

for seasonality reasons (more than 60% of produced ice-creams are sold between May

and August).

There are two main distribution channels: the retail and impulse. Through the retail

channel the products are delivered to supermarkets and hypermarkets. Even if direct-

to-store deliveries are possible, the products are usually delivered to a retailer

distribution center and then shipped to the points of sale. The impulse channel is used

to serve restaurants, cafés and special clients, e.g. theme parks. The ice-creams are

shipped to local distributors that can either buy the products (dealers) or simply

provide logistic services (distributors). Both channels end up refilling the freezers

located in the points of sale. In this paper, we will focus on the impulse channel,

150

which is considerably more complex than the retail one, from a temperature control

point of view.

3 Technological Solutions for the Cold Supply Chain

The most interesting technological solutions available to monitor the cold supply

chain are the Time Temperature Indicators (TTI), the Data Loggers, the semi-passive

RFId (Radio Frequency Identification) tags and the Wireless Sensor Networks

(WSN).

TTIs are smart labels that help verify whether perishable goods have been exposed

to harmful temperatures during their transit or storage. Placed directly on the

packaging of each single product, they provide an easy-to-read visual indication of the

accumulated time and temperature history of a product. [5]. In order to assure an

acceptable level of reliability, TTIs require a preliminary study on how the product

deteriorates and on how the TTI responds to a temperature change. Although applied

directly on product and so enabling continuous control of the cold chain, real-time

management is not possible, since actions can be taken only after the problem has

occurred. Moreover, the intensity of the problem is not measured, and a small break

of the cold chain will be reported in the same way as a severe one.

Data Loggers are probably the most widespread tool for temperature monitoring.

They are medium-sized electronic devices, with a separate power source (e.g. vehicle

battery) that collect and record data over time. Unlike the TTIs, the Data Loggers

support real-time response, and alarms can be generated. The main disadvantage is

that they monitor the temperature of air and not that of product, being positioned on

the containers. Moreover, they do not provide any information on what happens when

and where the Data Loggers are not present (e.g. loading/unloading activities).

Radio Frequency Identification (RFId) is an automatic identification technology

with which data can be stored and remotely retrieved data using small devices called

RFId smart tags. For temperature monitoring, semi-passive tags have to be used, with

a small battery on board to power the sensor, while the communication ability still

relies on passive RF data exchange. Their working principle is similar to that of a

Data Loggers but, being much smaller and power independent, they can be physically

put onto products: cost reasons make it impossible to tag each single product, but with

a careful selection of tag positioning a continuous monitoring capability can be

achieved. Moreover, they can use RF communication for a quick data download, even

if a specific hardware and software infrastructure has to be put in place to interrogate

the tags (with a reading distance of about 5-7 meters for UHF semi-passive tags); in

this regard, they do not allow for real time control, since they have to be interrogated

to communicate that a temperature shock has occurred [6].

Wireless Sensor Networks are mesh networks consisting of spatially-distributed

autonomous tags (nodes) endowed with sensors to monitor any physical parameters of

their environment cooperatively. Sensor nodes are very small computers consisting of

a processing unit with limited computational power and memory, one or more

sensors, a wireless communication device and a power source (e.g. a battery) [6]. As

for the RFId semi-passive tags, the nodes are very small, and can be put within the

shipped batch, even though not on each single product, and data can be automatically

151

recorded and downloaded. In this case the reading distance is much greater than

before (up to hundreds of meters) and the nodes can work as a mesh communication

network (so one node can act as a bridge for another node’s communication).

Therefore a much simple reading infrastructure is needed (usually, a final node with

gateway capabilities is used to link the mesh sensor network to an internet

connection). Moreover, being active, they can be used for real time control (alarm

issuing).

The short comparison in Table 1 allows to point out the most important

technological performance dimensions which characterize these solutions.

Table 1. Comparison of the four technological solutions in terms of performance.

TTI Data Lo

ers Semi-

assive RFId WSN

Realtime

No Yes No Yes

Resolution

Single Ite

Air

Precision

Bad Good Good Good

Data download

None Cable RF RF

Readin

rocess /

infrastructure

Manual Manual

Portals or handheld

readers, 5-7 meters

readin

distance

Wi-Fi access points,

hundreds meters reading

distance

Air / Selected items in the shipping lot

The technological solutions presented might yield different performances, but they

are also characterized by different costs and organizational complexity (cfr. Figure 2).

Technologies

Costs

RFIdsemi‐passivetags

WSN

DataLoggers

TTIs

Legenda:

Lowimpact

Mediumimpact

Highimpact

Verylowimpact

HWandSW

Installation

Training

Project

management

Project

management

Maintenance

Organizationalcomplexity

Implementation

difficulty

Implementation

difficulty

Cooperation

needs

Cooperation

needs

Trainingneeds

Changeinthe

waypeoplework

Changeinthe

waypeoplework

Information

sharing

Information

sharing

Fig. 2. Comparison of the four technological solutions in terms of cost and complexity.

TTIs do not require a dedicated infrastructure, and they are quite cheap if they are

applied at pallet or case level, while the cost dramatically increases if they are applied

to all items. This solution is quite easy to manage. The main difficulty concerns the

interpretation of the color in intermediate situations, which requires specific training.

Data Loggers require a slightly higher initial investment, depending on their exact

number. However, they can be used for a long time without requiring any expensive

maintenance, so the overall life-cycle costs are quite low. Furthermore, project

management and technology roll-out is quite easy.

The use of RFId semi-passive tags implies higher costs (for tags, readers, antennas,

middleware): such a solution means much greater organizational complexity,

152

especially when the tags are applied to the products. On the other hand the system is

very precise in monitoring the goods’ temperature.

The WSN is the most expensive solution, because of the high cost of the wireless

sensors, the bridges and the gateways used to download the data. The installation

phase may be complex and requires the collaboration of various actors but, once in

place, the operational activities are not affected; sensors are very versatile and the

quality and visibility (real-time internet browsing) of the collected data are excellent.

4 The Application of Wireless Sensor Networks in the Nestlé

Ice-Cream Division Italy

In this section, we focus on how WSN can improve the cold chain management in

Nestlé Ice-Cream Division Italy, with respect to the impulse channel (cfr. Section 2).

Central database

Plants First-tier

warehouses

Second-tier

warehouses

Cafè /

restaurant

Internet Internet Internet

WSN

GPRS GPRS GPRS

Fig. 3. The WSN solution to monitor Nestlé Italy ice-cream supply chain.

Sensors are installed in the cold rooms of the warehouses (plant, first-tier and

second-tier warehouses) and the sensed data is sent to the central database via the

Internet (cfr. Figure 3). During transport, measurement data is locally recorded and is

sent to the central database on arrival to a warehouse. Alternatively, the data might be

transferred in real-time using a GSM/GPRS connection. At the point of sale, the

sensors are installed in the freezers. Since cafés and restaurants usually do not have an

Internet connection, data is collected by the access boxes installed in the trucks. It is

in monitoring and automatically gathering the temperature data from the points of sale

that WSN reveal their potential in comparison to traditional solutions which offer

only local control. All information is globally accessible on the Web, once the trucks

have ended their route.

The system provides full traceability of the temperatures, of the alarms and of the

actions relating to the refrigeration assets.

5 Benefits for the Supply Chain

The WSN solution described above can offer various benefits in four main areas:

product quality, supply chain visibility, asset management and brand image.

153

5.1 Product Quality

The WSN helps maintain the right conservation and transportation conditions, thus

assuring better product quality. Should a problem occur, the system helps the

company to identify where and when the cold chain was broken. Every year, Nestlé

Ice-Cream Division Italy has to destroy a certain amount of products because of

breaks in its “internal” cold chain, i.e. before shipping from the first-tier warehouses

to the distributors. The continuous monitoring of the temperature across the whole

supply chain (both in fixed and mobile environments) will help identify out-of-control

situations, thus preventing breaks in the cold chain. The current costs might be

reduced by 30%.

However, most of the problems do not occur within the boundaries of the

company, but after the products have been shipped to the local distributors, since the

complexity is higher (e.g. higher number of stops per truck). Complaints from

customers (i.e. restaurants, the cafés) about poor product quality making products

difficult to sell are very frequent. As the company is currently unable to identify

responsibility of quality problems, Nestlé offers a block discount to all dealers.

Considering both the reduction in the number of cold chain breaks and the

objective assessment of the responsibilities, the WSN solution can yield a 50%

reduction in the cost of poor quality (about € 250,000/year).

5.2 Supply Chain Visibility

As described in the previous sections, the ice-cream supply chain is made up of

various actors organized in multiple tiers. Since supply chain performance

significantly depends on the behaviour of all the actors, it is important to monitor

effectively the whole supply chain. The WSN solution helps identify critical

situations: currently, people working in the quality function operate in two ways: they

react to a problem that has already occurred (e.g. a distributor affirms he has received

unsellable goods) or they carry out statistical quality controls. Thanks to the on-line

visibility of the thermal conditions in the whole supply chain it is possible to optimize

these activities, so deciding to carry out controls when and where problems are more

likely to occur. It is therefore possible to increase the productivity and even to reduce

the number of people needed to carry out the quality controls.

5.3 Asset Management

Since the freezers located in the cafés and restaurants are usually owned by the

producer, e.g. Nestlé Italy, one of the most critical issues in the ice-cream supply

chain is freezer management. Currently, visibility is very low which leads to an

increase in freezers turnover and to high maintenance and energy costs.

However, data on the temperature cycles inside the freezers would yield many

benefits. First, it would be possible to identify more easily the freezers that need to be

serviced. The maintenance costs are usually incurred by the local distributors, but

since the producer indirectly pays for this service they can both benefit from the WSN

solution. A prudent estimate of the reduction in the annual freezer maintenance costs

154

would be 10%, corresponding to €1.5-2/freezer. If we consider that a large ice-cream

manufacturer usually owns thousands of freezers, the order of magnitude of these

benefits becomes clear. For illustrative purposes, consider a manufacturer owning

100,000 freezers managed by 40 local distributors: thanks to the WSN solution, each

distributor can save about €4,000-4,500/year, leading to overall benefits of €150,000-

200,000/year.

Second, the point of sale could reduce its energy costs. A preliminary analysis of

the temperatures inside the freezers showed that they are often lower than necessary.

A more careful control of the temperature would facilitate the identification of these

cases. Assuming a 6% reduction in overall energy consumption (about €20/year for

each freezer), a substantial monetary benefit could be achieved (€2 million in the

previous example), in addition to the positive effect on the environment.

Third, it has been estimated that a better maintenance policy could increase the

freezer lifecycle by two years. Taking the figures of the previous example, the cost

saving might reach €700,000/year.

Finally, the proposed solution enhances freezer traceability. It would be easy to

know where they are and if they are used in the proper way. For instance, the

monitoring of the internal temperature prevents unwanted use of the freezers (e.g. to

store cakes, vegetables, ham, etc.).

5.4 Brand Image

Better control on the distribution process – both inside and outside the manufacturer

boundaries – could lead to a better brand image.What is, however, the consumers’

perception of this issue? In order to understand more fully consumer awareness and

the value they assigned to temperature monitoring, a descriptive survey has been

carried out, involving about 200 consumers selected among the people buying frozen

foods in five retail stores located in the north of Italy. To provide general relevance

the focus is not only on ice-creams, but on all products that have to be stored and

transported at controlled temperature, with a special attention to frozen goods.

The questionnaire was divided into four main sections. First, some data on the

interviewee (e.g. sex, age) and his/her shopping behaviour were collected. In the

second and third sections, the consumers’ knowledge of the cold chain was

investigated as well as their reaction to a break in the chain. Finally, the value

assigned by the consumers to temperature monitoring was explored.

The main results can be summarized as follows. About 50% of the interviewees

have a good knowledge of what the cold chain is and usually look at the conservation

temperature given on the product packaging. It is interesting to note that this

percentage is similar (48%) to that of the number of people experiencing a break in

the cold chain. Moreover, about 65% of the interviewees know that there is a

thermometer inside the freezers located in the point of sale. About 35% of the

interviewees think that if a break in the cold chain occurs, it is possible or even likely

that the product is sold anyway. In order to prevent such phenomena, more than 30%

of the interviewees declare they would accept an increase in price of 5% or even more

in order to reduce the likelihood of temperature breaks.

155

6 Conclusions and Managerial Implications

Careful study of the state-of-the-art technologies highlights the existence of

innovative solutions to improve cold chain management. In particular, the paper

investigates the benefits offered by Wireless Sensor Networks in terms of improved

quality, increased supply chain visibility, asset management and brand image. An

application to the Nestlé Ice-Cream Division Italy supply chain has been provided.

Future developments of the research must also investigate other technological

solutions providing a quantitative assessment of the costs and the benefits and

therefore a useful tool to support investment evaluation.

Even if the presented study deals mainly with the ice-cream supply chain, the key

principles and conclusions can be extended to other food cold chains.

Acknowledgements

This study has been carried out as part of the activities of the RFId Solution Center,

the competence center on RFId technologies of the Politecnico di Milano, HP and

Intel. We would like to thank IP01, a company specialized in WSN solutions, for its

support in the definition of the solution architecture.

References

1. RFId-IPO, Italian Permanent Observatory on RFId technologies: RFId: alla ricerca del

valore, research report - in Italian (2007)

2. Chappell, G., Durdan, D., Gilbert, G., Ginsburg, L., Smith, J., Tobolski J.: Auto-ID on

Delivery: The Value of Auto-ID Technology in the Retail Supply Chain, White Paper,

Auto-ID Center, MIT (2002)

3. Sellitto, C., Burgess, S., Hawking, P.: Information quality attributes associated with RFID-

derived benefits in the retail supply chain, International Journal of Retail & Distribution

Management, Vol. 35, No. 1 (2007) 69-87

4. Bogataj, M., Bogataj, L., Vodopivec, R.: Stability of perishable goods in cold logistic

chains, International Journal of Production Economics, Vol. 93-94 (2005) 345-356

5. Giannakourou M.C., Taoukis P.S.: Systematic application of Time Temperature Integrators

as tools for control of frozen vegetable quality, Journal of Food Science, Vol. 67, No. 6,

(2002) 2221-2228

6. Battezzati, L., Hygounet, J.L.: RFId - Identificazione Automatica in Radio frequenza:

Tecnologie e applicazioni, Ulrico Hoepli Milano (2003)

156