7UHQGVLQ)RRG6FLHQFH

7HFKQRORJ\²

Contents lists available at ScienceDirect

Trends in Food Science & Technology

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

Review

Global food security – Issues, challenges and technological solutions 7

a,∗ b c b d
Ultan Mc Carthy , Ismail Uysal , Ricardo Badia-Melis , Samuel Mercier , Colm O'Donnell ,
Anastasia Ktenioudakid
a
School of Science & Computing, Department of Science, Waterford Institute of Technology, Ireland
b
Department of Electrical Engineering, University of South Florida, Tampa, USA
c
Departamento de Ingeniería Agroforestal, ETSIAAB, Universidad Politécnica de Madrid (UPM), Edificio Motores, Avda. Complutense 3, 28040 Madrid, Spain
d
School of Biosystems & Food Engineering, University College Dublin, Ireland

A B S T R A C T

Background: Food security is both a complex and challenging issue to resolve as it cannot be characterized or
limited by geography nor defined by a single grouping, i.e., demography, education, geographic location or
income. Currently, approximately one billion people (16% of global population) suffer from chronic hunger in a
time when there is more than enough food to feed everyone on the planet. Therein lies the Food security
challenge to implement an ability to deal with increasing food shortages, caused by a combination of waste and
an ever expanding world population. At current levels prediction state that we must increase global food pro-
duction by 70% on already over exploited finite infrastructures before 2050.
Scope and approach: This review paper firstly introduces the concept of Food Security with an overview of its
scale and depth in the context of the global food industry. It then highlights the main sources. The readership is
then introduced to the key factors affecting food security and highlights the many national and international
measures adopted to tackle the problem at both policy and technological level.
Key findings and conclusions: Food experts indicate that no one single solution will provide a sustainable food
security solution into the future. Collective stakeholder engagement will prove essential in bringing about the
policy changes and investment reforms required to achieve a solution. Achieving truly sustainable global food
security will require a holistic systems-based approach, built on a combination of policy and technological
reform, which will utilize existing systems combined with state-of-the-art technologies, techniques and best
practices some of which are outlined herein.

1. Introduction access to food, (3) food utilization based on cultural and dietary re-
quirements and (4) food stability, i.e., the stability of its provision.
There is no shortage of definitions for food security available today. These pillars are represented in Fig. 1 and discussed in greater detail in
The FAO defines food security as when all people, at all times, have later sections. It is also evident that the onus is now on policymakers,
physical, social and economic access to sufficient, safe and nutritious governments, industrial practitioners, environmental non-for-profit or-
food, which meets their dietary needs and food preferences for an active ganizations and each individual to play their part in the food security
and healthy life. While there exist slight differences in the wording of its challenge to ensure that a high quality standard of food will remain
different definitions, the common underlying concept of food security is available in the foreseeable future.
that “all people at all times have access to sufficient, safe, nutritious food to The challenge of food security requires an ability to deal with in-
maintain a healthy and active life”. This requirement for food security is creasing food shortages for an ever expanding world population. With a
set in a reality where unstable food prices triggered by global scale predicted increase of 1.7 billion in world population between now and
events such as political instability, climate change and fuel shortages 2050, mankind is placing more and more pressure on the shrinking
have made the challenge of attaining and maintaining global food se- finite resources used to produce our food. The current model of an ever-
curity even more complex. The FAO/UNICEF have described food se- increasing population relying on finite resources is clearly un-sustain-
curity as a multi-layer concept focused on four key dimensions; (1) food able and increases the importance of ensuring that we strive for “re-
availability (2) food access, which includes physical and economical source efficiency” within a “circular economy”. A number of recent

∗
Corresponding author.
E-mail address: umccarthy@wit.ie (U. Mc Carthy).

https://doi.org/10.1016/j.tifs.2018.05.002
Received 4 August 2017; Received in revised form 8 December 2017; Accepted 1 May 2018
$YDLODEOHRQOLQH0D\
‹(OVHYLHU/WG$OOULJKWVUHVHUYHG
U. Mc Carthy et al. 7UHQGVLQ)RRG6FLHQFH 7HFKQRORJ\²

Fig. 1. The four Pillars of Food Security.

investigations indicate that anywhere between 30 and 50% of food groups and national strategies it is clearly evident that maintaining the
produced is never consumed and inevitably goes to waste (Gunders, “status-quo” in terms of simply increasing production to meet current
2012). In the US, each citizen wastes upwards of 400 pounds (approx. needs is no longer viable. This strategy is not applicable because the
180 Kgs) of food per year, while in Europe this figure stands at 173 Kg global population is expanding rapidly while our food is being pro-
of food wasted per year. The total cost associated with food waste for duced using shrinking natural resources. A total of 10 million ha of land
the EU-28 in 2012 was estimated at €143 billion. Similar statistics have is lost each year through soil erosion and a further 10 million ha due to
revealed that the US spends up to $218 billion per year (1.3% of GDP) irrigation related issues (Maggio, Criekinge, & Malingreau, 2015). The
on growing, processing and transporting food that is never eaten. In U.N. has predicted a 0.96% annual increase in the global population
Canada, it is estimated that food wasted annually is worth more than between 2015 and 2030 followed by a yearly on year increase of 0.63%
$25 billion, nearly 2% of Canada's gross domestic product (Young, between 2030 and 2050, resulting in a global population increase from
2012). its current 7.3 billion to 9 billion by 2050. This population growth is
The devastating impact food security has on humankind “the human expected to occur mostly in the lower income and less developed
effect” cannot be ignored and presents significant societal challenges countries, which traditionally face more significant food security issues
requiring immediate international attention. Latest estimates indicate compared to developed countries. Therein lies a requirement to adopt a
that approximately 795 million people in the world – just over one in more collaborative vision towards food security encompassing all sta-
nine – were undernourished in the years 2014–16 (FAO, 2015c) and an keholders both nationally and internationally. This will require, at the
estimated 805 million people were unable to access sufficient supplies very least, a clear and concise focus on areas such as infrastructure at all
of food between 2012 and 2014. The FAO recently reported that 60% of actor levels, communication (between all partners of the supply chain),
hungry people on the globe are women and almost 5 million children collective efforts, and clearly defined goals just to name a few. There
under the age of five die of malnutrition-related causes every year will also be a requirement to clearly focus on issues such as human
(Worldfoodday, 2016). While being a startling figure it should be trends, dietary requirements, urbanization, natural resources and cli-
highlighted that it represents an improvement of 209 million people mate change, all of which will be discussed with in more detail in later
compared to 1990 and 1992 (FAO, 2015b). The stark and disturbing sections.
reality behind the food security challenge includes the fact that an es-
timated 1 in 7 Americans are food insecure today (2016). Recent figures 2. Overview of current global food industry
in the EU report that in 2013, 55 million people (11%) reported
themselves as being unable to “afford a meal with meat, chicken, fish In an attempt to meet the ever changing nutritional requirements
(or vegetarian equivalent) every second day”. Today in 2017, 31.7% of and preferences of consumers across the globe, the global agri-food
SSA's (Sub-Saharan Africa) population are food insecure. Even more sector is in a continual state of flux and restructuring. Key challenges
staggering is that this figure will remain above 20% by 2027. Globally within this industry include an attempt to strike a trade-off between
the Asia region has the largest number of food-insecure people in 2017, product price versus safety, quality, variety and demand. To effectively
with 315.2 million individuals food insecure (ERS, 2017). This Asia meet these challenges and achieve cost efficiencies, industrial practi-
region is now the worlds fastest growing region whilst currently being tioners are required to address each of the major key areas in which a
home to 56% of the global population (Asian Development Bank, product is produced, processed, stored, distributed and accessed around
2013). The Asia region has the second largest food gap—10.8 million the globe. Similarly, in an attempt to achieve market success, recent
tons of grain in 2017, substantially below the 16.7 million tons for SSA. studies have reported that global food producing industries are now
This “food gap” measures the amount of food necessary to allow all adopting strategies aiming to gain a competitive advantage through
income groups to reach the caloric target. “category-focus ” as opposed to other industries where the key players
With respect to the numerous publications, whitepapers, action focus on portfolio management (in this $4 trillion per year industry)


U. Mc Carthy et al. 7UHQGVLQ)RRG6FLHQFH 7HFKQRORJ\²

(USDA, 2016). The adoption of a product focused strategy helps com- given that the vast majority of food supply chains span globally. When
panies becoming true leaders in their field for particular products and considering the supply chain one must also consider a number of linked
achieving global economies of scale. activities including sourcing of raw materials and parts, manufacturing
In the U.S., agriculture and agriculture-related industries con- and assembly, warehousing and inventory tracking, order entry and
tributed $835 billion to U.S. gross domestic product (GDP) in 2014. The order management, distribution across all channels and delivery to the
output of America's farms contributed $177.2 billion of this sum—- customer (Mc Carthy et al., 2012). To successfully reduce food waste
about 1 percent of GDP. In 2014 agriculture was responsible for the and increase resource efficiency across the farm-to-fork continuum one
employment of 17.3 million full- and part-time workers, about 9.3 must adopt a strategic view through combination of local (national) and
percent of total U.S. employment. Direct on-farm employment provided international perspectives – and key to achieving a global solution lies
over 2.6 million of these jobs whereas inter-related industries supported in the ability to successfully and holistically merge these local solutions.
an additional 14.7 million jobs. Food services and accounted for the Recent trends towards global trading and the formation of global
largest share—11.4 million jobs—and food/beverage manufacturing scale supply networks make the task of supply chain management more
supported 1.8 million jobs. In 2013, the U.S. food and beverage man- and more challenging and have significantly increased competitive
ufacturing sector employed about 1.5 million people, or just over 1 pressure. Not only do organizations have to continually evolve and
percent of all U.S. nonfarm employment (USDA, 2016). develop new strategies to meet the needs of their customers, they must
Similar trends can be observed for other countries. For example, in also develop parallel strategies to outperform their competitors to en-
2013, the agricultural and agri-food system generated $106.9 billion, sure sustainable corporate success. This reinforces the need for the
equaling 6.7% of Canada's overall GDP - a trend which has increased implementation of state of the art systems including a combination of
annually since 2007, the exception being the economic recession of wireless technologies, operating models, networking protocols, hard-
2009. Employment in the majority of industries in this sector continued ware (such as wireless sensor networks (WSN) and radio-frequency
on an upward trend and accounted for one in eight jobs in Canada, identification (RFID)) and software to ensure food is distributed in a
employing over 2.2 million people. The food service industry was the safe and secure manner (Badia-Melis et al., 2014).
largest employer in the AAFS, accounting for 5.3% of all Canadian jobs. Future developments must be built around sustainable food supply
On the other side of the Atlantic, between 2008 and 2010 the networks using innovative technologies. This can only be achieved in
European Union imported close to €60 billion worth of agricultural an organizational learning environment, through innovative value
products from developing countries annually. The food value chain in adding technologies that will improve the quality of our lives with a
Europe generates added value of €800 billion and a turnover of €4 negligible cost to the environment or to food security (McCarthy et al.,
trillion. Employment stands at 46 million people in more than 15 mil- 2013). As previously mentioned current supply chain challenges must
lion holdings or enterprises in agriculture, the food industry, and food be addressed both nationally and internationally given that in many
trade and services. The food and beverages industry is EU's largest countries, transport and logistics costs can be as high as 20%–60% of
manufacturing sector, in terms of turnover (€1.2 billion, or 1.8% of EU the food selling price (OECD, 2013a). From a trading and purchasing
Gross Value Added - GVA), employment (4.22 million jobs), value perspective it is difficult for a manufacturer to export at a competitive
added (€ 206 billion, or 12.8% of EU manufacturing) and exports (€ 92 price or import at a competitive cost if the transport and logistics sector
billion, or 18% of EU exports). Small and Medium Enterprise (SME) is volatile and/or dysfunctional (OECD, 2013a).
companies account for 99.1% of the sector. Internationally, recent reports have highlighted that, in some cases,
Irrespective of the geographic location, much emphasis has been a single day decrease in time spent at sea can increase trade costs to the
placed in the role of the family farmer in solving world food hunger. tune of 4.5% (OECD, 2013b). An investigation within the U.S. reported
There are up to 500 million family farms globally which accounts for that reducing shipping times by a single day can yield a 0.8% ad-va-
98% of farming holdings worldwide. Typical family farms include fruit lorem for manufactured goods. Furthermore, this same single day re-
and vegetable farms, grain farms, orchards, livestock ranches, and even duction will also increase the probability (by a factor of 1–1.5%) that
fisheries and those that harvest non-wood forest products. In Brazil the U.S. will import the given product from that country. Based on data
family farms provide up to 40% of the major crops and in the U.S. these from 133 countries' trade facilitation performance (i.e. border stream-
farms are responsible for producing 84% of all produce. Such farmers lining) can be responsible for approximately 14% of the variance in
have an intimate knowledge of their land: its history, needs and pro- bilateral trade costs across trading partners (OECD, 2013a). Similarly, it
ductive capacity and are viewed by many as being custodians of the has been shown that a single extra day spent in customs causes a 2.8%
land as opposed to exploiters (FAO, 2014). The figures discussed in this decline in growth rates of exports for freight in transit (Martincus &
section serve to remind the readers of the importance of the Global Graziano, 2012). These figures confirm the importance of efficient
Food sector and security and the impact it has on major global econo- supply chain management at the international level. At national level, it
mies. is expected that in countries where there is poor internal connectivity
(physical and virtual), the direct and indirect costs associated with food
3. Global food supply chains transport are unnecessarily high when set against the aforementioned
20–60% of food prices being governed by transport and logistics costs
There are a variety of definitions for “supply chain management”, (World-Bank, 2012). In countries with poor internal connectivity,
yet their common aim is to coordinate and integrate all activities re- where transport and logistics services are under-developed, and supply
lating to a product during processing and transport. Nowadays, food chain governance is poor, both the direct and indirect costs of trans-
chains have evolved to become highly distributed, heterogeneous, co- porting food from the farm to the consumer are significantly higher
operative and globalized processes with extremely diverse require- when compared to countries with developed connectivity. This leads to
ments, which are in many cases dictated by the product and its eventual inevitably higher consumer prices (OECD, 2013a). Transport and lo-
destination (Badia-Melis, Villalba, & Barreiro, 2014). Global food gistics services also affect a country's ability to respond to market price
supply chains represent delicate balances between the transportation of shocks. Lower transport costs can decrease the selling price of imported
products across the globe while factoring internal considerations such goods and ensure that a greater share of the selling price of exported
as prevalence of food spoilage, remaining shelf life and an array of goods is given to the producer (OECD, 2013a). Recent OECD research
external factors such as cost and distance-to-market. The application of has reported that a 10% improvement in transport and trade-related
appropriate processing of food is of critical importance for achieving infrastructure quality can increase developing countries' agricultural
sustainable supply. Nevertheless, food transport and distribution cannot exports by as much as 30% (OECD, 2013a).
be ignored when considering food security and sustainability especially Irrespective of a national or international approach, it is also


U. Mc Carthy et al. 7UHQGVLQ)RRG6FLHQFH 7HFKQRORJ\²

important to remain cognizant of social movements and population technical limitations in harvesting techniques, storage and cooling fa-
shifts which have created challenges related to consumer demand for cilities in challenging climates, infrastructure, packaging and marketing
specific food regardless of seasonality or geographical location. It is also systems (FAO, 2011). This may not necessarily be the case in developed
worth highlighting that efficient food supply chains have the potential countries where the main causes of waste can all too often be linked to
to reduce food waste as well as the cost to the consumer given that the purchasing patterns and education therefore supporting that both local
cost of food waste is increasing towards the consumer end of the supply and international perspectives are required for a solution. The net result
chain. It is now clear that improvements in global supply chain man- is that full stakeholder engagement is critically important in delivering
agement may only be achieved through the implementation of chain a sustainable solution.
wide monitoring systems that facilitate the exchange of information in
unison with the product. Such systems will increase the transparency 5. Factors affecting food security
and security of the food supply resulting in more flexible and responsive
global supply chains and a reduction in global food waste. The EU has revealed that the food and drink industry is facing a
decrease in competitiveness which may be linked to a lack of trans-
4. Sources and causes of global food waste parency in the food supply chain, sub-optimal business-to-business re-
lationships, a lack of attractiveness for skilled workers, and a lack of
There are variations in the reported amounts of food wasted glob- market integration across EU countries (EU, 2016). These issues are not
ally. The U.N. FAO estimates that each year, approximately one-third of exclusive to the EU and can be directly attributed to food security
all food produced for human consumption in the world is lost or wasted. challenges being experienced across other geographical regions. While
Similar reports have documented food waste figures as being as high as it has been presented thus far that there is no single solution for solving
50%. The U.N. Environment Program says that 222 millions of tons are global food security, it can also be deduced that there is no single root
annually wasted in developed countries, almost as much as the entire cause for the issue. This section outlines some of the major causes of
production of the sub-Saharan Africa (230 million tons). A study per- global food insecurity which should be taken into consideration to
formed in U.S., Canada, Australia and New Zealand reveals that during identify the critical factors towards achieving a solution. To address
production, 20% of the fruit and vegetables are lost; consumers waste these issues correctly the authors focus on the definition of food se-
up to 28%, and 12% is wasted during distribution mainly because of the curity according to the FAO as a time “when all people, at all times, have
lack of refrigeration control (Gunders, 2012). Irrespective of the physical, social and economic access to sufficient, safe and nutritious food
aforementioned reports being taken under consideration it is clearly which meets their dietary needs and food preferences for an active and
evident that global food waste is at unacceptable levels. healthy life”. The concept of food security is built on 4 key pillars, each
Recent publications have identified that up to 70–80% of food waste with its own core focus (Fig. 1). Each pillar is of equal importance and
occurs at customer facing businesses and households in the EU and U.S., must be treated with equal levels of consideration. These four key pil-
respectively. The total cost associated with food waste for the EU-28 in lars are (Aborisade & Bach, 2014).
2012 was estimated at €143 billion. Similar statistics have revealed that
the U.S. spends up to $218 billion per year (1.3% of GDP) on growing, i. Food Availability. This pillar entails having an adequate quantity
processing and transporting food that is never eaten. This equates to of food available at minimal notice. This pillar considers food that is
approximately 400 pounds (180 kg) of food wasted per year for every either produced locally by Food Business Operators or imported via
American citizen. In Europe this figure stands at 173 kg of food wasted the importing capacity of the region. It may also consider existing
per year (compared to a total of 865 kg produced). In addition to this food stocks within a particular region and also the provision for
critical statistic it is worth highlighting that up to 10 million tons of food aid that have been assigned from other countries. It is also
products are unsold and eventually wasted each year at farms and important to note that high food availability is not in itself sufficient
packing facilities due to aesthetic rejection (consumer will reject based to ensure food security (FAO, 2015a).
on the appearance). Similarly, in terms of waste recovery it has been ii. Food Access. This pillar considers both the physical and/or eco-
reported that less than 10% of waste is recycled at consumer facing nomic access to food. Factors of importance here include the pur-
businesses, whereas for manufacturers approximately 95% of industrial chasing power of a particular region and the levels of income (per
food waste is recycled, primarily for animal feeding purposes. These population) relative to other regions. Other factors for consideration
statistics lead the authors to assert the point that the introduction of include local infrastructures, such as transport modalities and or
financial implications for volume of wasted food could possibly reduce financial means to support trade within/across a particular region.
the overall food waste levels (Vared Sarah, 2016). Even though the cost iii. Food Utilization. This pillar focuses on the entire supply chain and
of food waste is higher at the consumer end of the supply chain, the deals with the way in which the food is handled from a safety
irony is that this is where the majority of waste occurs (Gunders, 2012) perspective. It focuses on hygiene across primary production, sec-
(Maggio et al., 2015). When considering food waste it is important to ondary processing, distribution, retail and household. This requires
consider the many factors which also have an impact on this issue. the creation of efficient and safe usage of food and also the creation
Factors include the different stages of the food product life cycle, which of nutritional awareness across stakeholders to ensure that they
include primary production, processing, wholesale and logistics, retail become more nutritionally aware of different foods and their re-
and markets, food services and household consumption. Fig. 2 presents spective benefits.
a breakdown of the sources of food waste across the value chain from iv. Food Stability. This pillar focuses on stability mainly from a supply
both a European and US perspective. and access perspective. Common factors with the potential to im-
It is clear that food waste in the world is at unacceptable levels, pact food stability include price, political stability and the local
however, as can be seen from Fig. 2 it is also important to note that the economy. Other factors which cannot be ignored include weather
volume and stage of supply at which the food is wasted also varies patterns with a negative impact on crop yields. It is generally ac-
according to the food type and geographical location. To facilitate the cepted that the poorer the nation, household or consumer, the more
adoption of a systematic global approach towards a solution, one susceptible it is to external factors (FAO, 2015a).
should be open to the idea that the issues of waste in one geographical
region or sector may already be solved in another geographical area. As previously mentioned there is no single cause or solution to the
Nevertheless, each demographical factor would present its own chal- issue of food security. The following section will outline some of the
lenges as well. For instance, the causes of food losses and waste in low- wider cross cutting causes and/or factors affecting global food in-
income countries are mainly connected to financial, managerial and security.


U. Mc Carthy et al. 7UHQGVLQ)RRG6FLHQFH 7HFKQRORJ\²

Fig. 2. Main sources of food waste in the EU and US.

Population Increase: As previously mentioned, the U.N. has pre- supply and demand perspective at global level as a result of inconsistent
dicted an 0.96% annual increase in the global population between now supply channels to market.
and 2030 and thereafter a yearly increase of 0.63% through 2050 re- Food versus fuel debate: Stakeholders have major concerns that
sulting in an overall global population increase from its current 7.3 food supply chains which have been traditionally destined for food
billion (UNFPA) to 9 billion in 2050. This population growth is ex- production may become redirected towards the production of biofuels
pected to occur mostly in the lower income, developing countries, due to economic pressure. Traditional crops which have been redirected
which traditionally have faced more significant food security issues for the production of biofuels include Maize, Oilseed and Sugar cane.
than developed countries. This can significantly decrease access to locally produced foods and can
Urbanization: By 2050 it is expected that more than 65% of the result in local populations not being able to afford locally produced
population will be living in urban areas. This trend of urbanization will food (Tenenbaum, 2008).
occur mainly in low-income countries and will be a key driving force Infrastructure: The issue of infrastructure spans the complete
behind the creation of dense demand pockets throughout the world supply chain and is relevant to all stakeholders. At primary production
placing increased demand on food supply chains. in more developed countries certain products are grown in artificially
Dietary demands: An increase in global agricultural and food created environments and require a high degree of control and mon-
products of 50% by 2030 and 110% by 2050 will be required to meet itoring while other foods are grown in the traditional farm setting and
the demands of growing population and rising incomes. As an example, at the mercy of the weather during that season. At the secondary pro-
it has been predicted that there will be a 40% increase in demand (kg cessing level the more developed countries have the ability to automate
per person per year) for meat in higher income countries and a 69% and modernize the process or the environment to increase production
increase in lower income countries between 2015 and 2050. Similarly, and in many cases achieve the required scale. Again, many regions do
the demand for dairy products is expected to increase by up to 70% not have highly automated production systems and rely on human ef-
between 2000 and 2050 (Maggio et al., 2015). Globally, the average fort. At the consumer level the more developed countries can generally
annual increase of total food production between 1961 and 2011 was offer the consumer products in finely presented and specific packages
122 MMT (Million Metric Tonnes) (Dou et al., 2016). containing marketing and nutritional information, which also help to
Natural resources: This is collectively considered under a combi- extend the shelf life of the food. In less developed countries this is not
nation of urbanization, poor water management and poor soil condi- always the case and products are very much at the mercy of the im-
tions, all of which are major factors contributing to soil becoming unfit mediate surrounding in which they are transported and stored. It has
for agriculture. Similarly we cannot ignore other critical natural re- been reported that is the same level of refrigeration used in developed
sources such as energy utilization and fuel dependence, which also countries were to be applied to developing growths this would result in
directly have the potential to undermine global food security if not those countries saving approximately 200 million tons (14% of the
managed and monitored correctly. We should aim to exploit renewables countries consumption) (Mercier, Villeneuve, Mondor, & Uysal, 2017).
where possible to reduce out carbon output. It has been reported today, As previously mentioned 31.7 percent of SSA's (Sub-Saharan Africa)
humankind use roughly one-half of the planet's vegetated land to grow population, approximately 301 million people, are food insecure in
food (WRI, 2013). Similarly, it is estimated that approximately 10 2017 which can be attributed, in part, to weak currencies and/or dis-
Million ha of land is lost each year through soil erosion and a further 10 ruptions along the food supply chain (ERS, 2017).
million ha due to irrigation related issues (Maggio et al., 2015). Water is Communication: Communication across all supply chain stake-
currently over-exploited and not treated as a valuable resource in the holders is critical. It is well-established that an ability to create in-
agri-food industry which is responsible for 70% of global fresh water formation sharing channels across stakeholders can significantly in-
consumption. In practical terms, the NRDC has reported that the vo- crease profits and reduce food waste (Mcmurray et al., 2013) (Mc
lume of water consumed by running a domestic shower for 104 min is Carthy et al., 2012). These information sharing channels improve
equal to the water required to produce a pound of chicken. Similarly the transparency across trading partners and increase the amount of in-
production of 1 pound (0.45 Kgs) of tomatoes, bananas, white rice and formation available on the arrival date of particular goods, such as
beef requires 5, 42, 60 and 370 min, respectively, of running water from information on its remaining shelf life and also significantly improve
a domestic shower (NRDC, 2016). warehouse management and logistical efficiency (Hertog, Uysal, Mc
Climate change: It is of common knowledge that local climate Carthy, Verlinden, & Nicolai Bart, 2014).
variations are exploited to promote cultivation of certain products.
Climate shifts and changes in weather patterns have drastic effects on
the productivity and economic prosperity at regional level and may in 6. Future food security - global action plans and visions
turn lead to population displacement and resource depletion. This in
turn has a significantly negative effect on food security from a both a As a general rule of thumb a simple, naive and all too often adopted
solution to food security would involve increasing domestic production


U. Mc Carthy et al. 7UHQGVLQ)RRG6FLHQFH 7HFKQRORJ\²

faster than the population growth for a particular area. Once these
criteria are met at global level the world will become food secure. This
however is a clear over-simplification of the issue and completely ne-
gates the reality of the situation. This section will outline in more detail
the current and future action plans to tackle the issue of food security.
There have been many action plans commissioned over the years both
nationally and internationally attempting to tackle the issue of global
food security and waste with varying levels of success. To successfully
increase food security and reduce food waste it is important to combine
a strategic multi-stakeholder approach with a deep comprehension of
the scale, source and causes of food waste. Similarly this approach
would have the flexibility to adapt to local issues while successfully
integrating an international strategy.
The European commission has recently adopted a resolution to de-
crease food waste by 50% by 2025. Similarly, in 2015 the US had
adopted a food waste reduction goal by the year 2030 under the ReFED
Fig. 3. Food security framework.
initiative (Vared Sarah, 2016). All these targets are being put in place to
promote what has been branded as a circular economy. With respect to
food security a circular economy means “an economy where the value of important to note that traceability needs both “tracing” and “tracking”
products, materials and resources are maintained for as long as possible with capabilities. According to (Petersen & Green, 2005), tracing is the
the aim of reducing waste and increasing a sustainable, low carbon, resource backward process where the product's origin is identified by history
efficient and competitive economy” (EU, 2015). and/or stored records in supply chain and tracking is the forward
Regardless of the strategy or scale of the adoption plan many visions process where the end users and trading partners are identified by lo-
for global food security require a number of core fundamental pillars on cation in supply chain. Tracking requires the ability to monitor the
which a wider and more comprehensive vision are actioned. Given that location of an asset in real or near real-time (McCarthy, Ayalew, Butler,
the solution is not a simple question of increasing production to meet Mcdonnell, & Ward, 2011). Collectively, both elements provide the
demand, it is recommended that each of these strategies adopt a re- basis of a successful traceability system (Van Dorp, 2002). To function
source-based perspective prior to their implementation. This resource- correctly and efficiently, a traceability system must also be both scal-
based perspective should include a complete evaluation of existing able and standardized. The system must be scalable in order to meet the
processes with the aim of improvement prior to the step by step in- requirements of the modern day global consumer and also standardized
cremental introduction of new systems. It has been reported that if the to ensure that the items being tracked do not become hindered by
US could reduce its waste by 20% - the net saving would be equal to two challenges including language barriers and or counterfeit product.
years of growth in productivity, where an average 7.7 MMT has been The unfortunate reality is that current traceability systems are
reported between the years 1961 and 2011 (Dou et al., 2016). With this characterized by their inability to link food chains records, information
in mind the criteria (perspectives) under which existing systems must inaccuracy and errors in records, which delays the transmission of es-
be examined include: sential information during time of food recalls (Badia-Melis, Ruiz-
Resource perspective: This requires a focus on the long-term Garcia, Garcia-Hierro, & Villalba, 2015). Effective traceability systems
preservation of the natural resources on which the food is cultivated. should have the ability to address the recall by firstly identifying po-
This will in turn promote the adequate preservation of the agricultural tentially at risk products and then withdrawing the unsafe ones. Such
assets (land and housing) in the best way possible to ensure maximum systems therefore need to be flexible and responsive. While being es-
profitability and sustainability. sential in the battle to attain food security and for consumer safety, food
Process perspective: The transformation of existing systems at traceability systems are not always off the shelf solutions and all too
each stage of primary production, secondary processing, distribution often require a level of customization. This adopted customization en-
and retail. This will often require financial initiatives to encourage re- sures these systems meet the requirements of the individual customer or
search, development and Innovation at regional and international level. the specific requirements (legal and otherwise) of the supply chain and
An additional element included in this perspective is stakeholder of the product (e.g., perishable, bulk, seasonal). To reinforce the busi-
training to promote resource efficiency at each stage. ness case for traceability systems, it is not only important to view these
Communication: This involves the implementation of a method/ systems as a means of meeting safety regulations; these systems have
system of chain wide communication across product/process stake- many additional applications. Traceability systems can help overcome
holders. Ideally such a communication flow will span the complete many issues such as food crisis management, quality concerns and
farm-to-fork continuum. Information sharing across the product value identity preservation concerns, fraud prevention, anti-counterfeiting
chain will promote global demand-driven food trading systems and (Dabbene, Gay, & Tortia, 2014) and food adulteration (Spink, Moyer, &
over time will facilitate a net balance between production and con- Speier-Pero, 2016).
sumption.
The above criteria should help stakeholders to develop a framework 6.2. Implementing monitoring technologies and techniques
on which strategies for improving food security can be implemented at
both local and international levels. Fig. 3 summarizes the requirements A core requirement inherent to the food security challenge is the
for the development of a Food Security Framework as outlined above. ability to secure a full product chain-of-custody. This requires the
Generic strategies and areas of development in recent times with ability to monitor/query the state of the product during transit. These
respect to food security include, but are not limited to the following. technologies are generally autonomous, of small dimensions and record
a number of product's critical parameters including temperature and
6.1. Standardized food traceability systems relative humidity. Recent developments have increased the monitoring
capabilities of sensors, notably regarding the measurement of the con-
Traceability is defined as “the ability to access any or all information centration of gases, presence of pathogens, leaf freshness, mineral de-
relating to that which is under consideration, throughout its entire life cycle, ficiency all of which combine to predict the remaining shelf-life of the
by means of recorded identifications” (Olsen & Borit, 2013). It is product. Many of these sensor technologies are in their infancy however


U. Mc Carthy et al. 7UHQGVLQ)RRG6FLHQFH 7HFKQRORJ\²

a significant number are currently being commercially deployed. reduce food waste and energy consumption along the supply chain
Monitoring food temperature accurately to ensure the longest shelf life (Gwanpua et al., 2015). CanGRASP is a GIS-based simulation tool used
represents a difficult challenge due to the number of factors involved. for the traceability of contaminated food during foodborne outbreaks,
Investigating the temperature gradient data inside refrigeration rooms, the development of mitigation measures to prevent outbreaks and the
warehouses, containers and trucks is a primary concern for the in- training of stakeholders involved in public health risk assessment
dustry. Any temperature disturbance can undermine the efforts of the (Hashemi Beni et al., 2012; LeBlanc et al., 2015). The combination of
whole supply chain (Mahajan, Caleb, Singh, Watkins, & Geyer, 2014). efficient wireless temperature monitoring systems with recent software
Inadequate temperature is the second prevalent factor causing food- developments could open exciting new perspectives towards an im-
borne illnesses, surpassed only by the presence of initial microflora in provement at a global scale of food safety and, by extension, food se-
foods (Sánchez-López & Kim, 2008). curity.
High temperatures trigger the growth of pathogens and accelerate In combination these monitoring techniques and technologies offer
natural processes causing food decay (Giannakourou & Taoukis, 2003; all stakeholders enhanced transparency across the complete food supply
Gwanpua et al., 2015; Hertog et al., 2014). Low temperature can also be network and it is through the provision of this actionable data that
detrimental to the shelf life and the quality of perishable food because more informed decisions can be made which will aid in addressing food
of cold injuries (Aghdam & Bodbodak, 2014; Heap, 2006). Time-tem- losses and increasing food security.
perature profiles measured along supply chains in Canada, France and
Greece have showed that the temperature of perishable food frequently 6.3. Internet-of-things/big data and global food safety
increases above the desired limit (Derens, Palagos, & Guilpart, 2006;
Koutsoumanis, Pavlis, Nychas, & Xanthiakos, 2010; McKellar, Leblanc, Recent years have seen groundbreaking advances in wireless sensor
Rodríguez, & Delaquis, 2014). Notable causes explaining the increases and data processing technologies paving the way for a global scale
of temperature above the desired limit include the inappropriate pre- cyber-physical infrastructure, which we now term the “Internet-of-
cooling of the food, poor performance of temperature control systems, Things” (IoT). Similarly, substantial amounts of data generated by IoT
temperature oscillations caused by the on-off cycles of the refrigeration implementations and their subsequent analytics, such as advanced
systems, local heat sources in trucks or warehouses, temperature abuses statistics and machine learning, to create actionable information are all
during truck loading and unloading, overloading of refrigerated display included under the term “Big Data”. However, the wide-scale adoption
cabinets, temperature abuses during transportation by the consumer of intelligent information systems still represents a particularly elusive
and high temperatures of domestic refrigerators (Carullo, Corbellini, problem specifically for the global food industry. In fact, as the world's
Parvis, & Vallan, 2009; Foster, Swain, Barrett, & James, 2003; top economy, in the United States alone, up to 40% of all produced food
Jedermann et al., 2011; LeBlanc et al., 2015). is inconceivably wasted as 50 million Americans continue to live in
Approximately 12% of food wasted in the USA occurs during dis- food insecure households (Gunders, 2012) (DeNavas-Walt & Proctor,
tribution, mainly because of inappropriate storage temperatures 2014).
(Gunders, 2012). The amount of food wasted because of inappropriate IoT and data analytics or more specifically machine learning – when
refrigeration is even more important in developing countries, because considered individually neither concept is brand new. Machine learning
of the absence of proper refrigeration equipment, high energy cost of has been formally defined in the literature since 1980s, IoT is, at its
refrigeration and lack of knowledge on the impact of temperature on core, a modern and holistic redescription of Internet connected devices
food safety. It is estimated that if developing countries could apply the and cyber-physical systems (Michalski, 1983; Weiser, 1999). In fact,
same level of refrigeration that is used in developed countries, more considered as one of the “founding fathers” of machine learning, Mi-
than 200 million tons of perishable food waste would be avoided per chalski initially worked on computer algorithms to recognize hand-
year, approximately 14% of their yearly consumption (IIR, 2009). En- writing as early as 1960s, and wrote the first volume of his landmark
suring the proper refrigeration of perishable food at a global scale and work “A Theory and Methodology of Inductive Learning” in 1983 which
throughout the entire supply chain should represent a key objective in developed the framework for many more algorithms to follow
our endeavour to improve food security. Monitoring devices are used to (Michalski, 1983) (Vapnik, 1998). Building on Michalski's research
ensure temperature integrity; however resource limitations and cost Vapnik introduced data centered supervised learning on statistical
factors severely limit their use to one-per-pallet or even one-per-con- principles in his book titled “Statistical Learning Theory”. Since then, a
tainer scenarios (Badia-Melis, Brecht, Lowe, & Uysal, 2013). myriad of machine learning algorithms and principles have followed
Wireless technologies have been considered key technological en- mainly divided into four main categories: i) supervised learning which
ablers to promote supply chain transparency across all actors. The po- assumes completely labeled data in terms of input-output pairs to train
tential for monitoring technologies, WSN and RFID, has been suggested an individual algorithm, ii) unsupervised learning which tries to find
by several studies and extensive literature has been published to ad- structures in unlabeled data, iii) semi-supervised learning operating on
dress the issue (Badia-Melis et al., 2014; Jedermann, Pötsch, & Lloyd, a mix of labeled and unlabeled data, and finally iv) reinforcement
2014b; McCarthy et al., 2011; Qi, Xu, Fu, Mira, & Zhang, 2014). Their learning which is based upon the idea to maximize a reward function by
adoption, to monitor the temperature of food along the supply chain optimizing specific actions in a given parametric setting (Mohri,
could significantly help improve food safety. The remaining shelf life of Rostamizadeh, & Talwalkar, 2013). Popular traditional algorithms in-
perishable food products can be estimated accurately from their time- clude artificial neural networks (ANN), k-nearest neighbor classifiers
temperature history using appropriate safety and quality models. From (kNN), support vector machines (SVM), classifier ensembles, Bayesian
the knowledge of the remaining shelf life of perishable food, First Ex- networks, and hidden Markov models (HMM) among many others
pired, First Out inventory management systems and dynamic expiry (Cortes & Vapnik, 1995; Dudani, 1976; Friedman, Geiger, &
date management systems can be implemented, both of which have the Goldszmidt, 1997; McClelland & Rumelhart, 1988; Rabiner, 1989;
potential to significantly reduce food waste (Tromp, Rijgersberg, Zhou, Wu, & Tang, 2002). In recent years, the concept of “deep
Pereira da Silva, & Bartels, 2012; do Nascimento Nunes, Nicometo, learning” has gained significant traction, which attempts to find ab-
Emond, Melis, & Uysal, 2014). stract structures in data without relying on explicit feature extraction
Time-temperature data can also be combined with recently devel- common to many supervised learning schemes (Deng & Yu, 2014). As
oped software such as the FRISBEE and CanGRASP tools. The FRISBEE more, or “big” data has become available, it has become practically
tool includes extensive databases of time-temperature histories mea- more feasible to successfully apply algorithms such as convolutional
sured along the supply chain and quality models predicting the re- deep belief neural networks, deep Boltzmann machines and sparse auto-
maining shelf life of perishable food, as well as optimization modules to encoders for a wide range of applications (Coates, Lee, & Ng, 2011;


U. Mc Carthy et al. 7UHQGVLQ)RRG6FLHQFH 7HFKQRORJ\²

Hinton, Osindero, & Teh, 2006; Lee, Grosse, Ranganath, & Ng, 2009; insurmountable barrier for many packers/shippers/retailers. Recent
Salakhutdinov & Hinton, 2009). advances in machine learning can help surpass this limitation by
Similarly, IoT concept, although gaining mainstream attention only achieving reliable and complete temperature mapping inside, for in-
recently, was introduced as far back as 1991 in Weiser's article “The stance, a refrigerated sea container using only a single container sensor
Computer for 21st Century” where he famously claimed “Specialized (Badia-Melis, Mc Carthy, & Uysal, 2016). However, as future direction,
elements of hardware and software, connected by wires, radio waves more research needs to be conducted to i) generalize such findings to
and infrared, will be so ubiquitous that no one will notice their pre- different types of containers and ii) more importantly study different
sence.” (Weiser, 1999). Considered by some as a mere redescription of types of produce to improve the robustness of the estimation models to
computing devices connected to the Internet, IoT examples are wide- a level which is acceptable by the industry as a whole.
spread (Li, Xu, & Zhao, 2015). For instance, a wireless sensor network
can easily be considered as an IoT framework as long as at least one 7. Conclusion
sensor node has access to a remote server (Dargie & Poellabauer, 2010).
Recent developments in radio frequency identification (RFID) domain To attain and sustain a truly global solution for food insecurity is no
and its wide-scale adoption in retail and supply chain can easily be simple endeavor, which requires a deep consideration for the concerns
considered as modern applications of IoT where Internet connected of a broad spectrum of stakeholders with an ability to merge both ex-
RFID readers communicate with RFID tags with unique identifiers at- perienced and predicted consequences. Stakeholders include farmers,
tached to objects in the environment (Welbourne et al., 2009). Some- processors, producers, distributers, wholesalers, retailers, consumers,
times defined as Web 3.0, which signifies the next generation of ubi- governments, environmental groups and a plethora of companies sup-
quitous computing web, IoT presents remarkable challenges and plying goods and services to all of them – all of whom will be re-
opportunities at the same time (Gubbi, Buyya, Marusic, & Palaniswami, sponsible for bringing about the policy changes and investment reforms
2013). required to achieve a solution. Hence, therein lies a requirement to
Many of these technologies are currently being adopted in the appreciate the scale of the issue prior to developing strategies. In es-
agriculture and food sectors globally. One of the key drivers of this is sence it is important to note that national or even local efforts, although
based on the aforementioned need to increase global food production being technically sound, may not have enough potential to create sig-
on an ever shrinking finite resource. Initially this started with enhanced nificant impact globally. In developing sure strategic aspirations one
automation, robotics, vision systems and in/on line analytical tech- must adopt a fully coordinated global effort to achieve sustainable food
nologies being incorporated to enhance production visibility and in- security.
crease operational competitiveness. Such technologies and systems are Equally as important is to successfully merge state-of-the-art re-
now being deployed across all sectors of the Agri-food sector globally. It search with informed high-level policy across a diverse range of po-
is now common place to deploy a suite of sensors across produce sites pulations and food types. From the complex issues surrounding food
and place sensors on animals monitoring a vast array of event for security presented herein it is important to remain open to the fact that
analysis, forecasting, and planning (Adrian, Norwood, & Mask, 2005) a sustainable solution to food security will facilitate the improvement
(Stephens, Jones, & Parsons, 2018). and transformation of food supply systems and will provide a level of
Similarly whereas advanced analytics and machine learning has enhanced transparency (RFID, WSN, IoT), never before experienced in
been used in the past for general supply chain studies such as demand the sector. It will also provide a provision of actionable data (IoT, Big
forecasting, or automatic setup of supply chain network, its applications Data) on which more informed decision can be actioned to improve
specific to cold supply chains have unfortunately been very limited both business and planning decisions.
(Carbonneau, Laframboise, & Vahidov, 2008) (Piramuthu, 2005). Again This must be set within an economic reality, capable of solving a
considering the fact that in the United States alone, more than 40% of global issue while also being designed to deliver changes locally. It is
all produced food is never even consumed; there is a significant margin obvious that a mere increase in production to meet demand is no longer
for improving sustainability and food (Gunders, 2012). Majority of data sufficient and a long-term perspective must be adopted and should
analytics studies and wireless sensor applications surrounding food consider both environmental and social perspectives as well as its fi-
supply chain focus on transportation, storage and distribution of per- nancial implications.
ishable food items – commonly called as the cold chain (Regattieri, For any large scale challenge, it is important to gain a complete
Gamberi, & Manzini, 2007, do Nascimento Nunes et al., 2014; understanding of the issue and underlying causes prior to attempting to
Jedermann, Behrens, Westphal, & Lang, 2006; Abad et al., 2009; Hertog implement a solution – and food security is no exception. One proposed
et al., 2014). Researchers have proven time and again the incredible and widely appraised solution for food insecurity is found in the ability
potential of reducing food losses by intelligent logistics such as shelf-life to correctly address the four pillars, food availability, access, utilization
prediction and using first-expired-first-out (FEFO) instead of industry and stability, and each of their specific issues. A diverse portfolio of
standard first-in-first-out (FIFO) (Hertog et al., 2014; Jedermann, challenges one might expect would be weather extremes including
Nicometo, Uysal, & Lang, 2014a). This is made possible through a un- flooding and droughts, political instability leading to supply issues and
ique combination of IoT frameworks such as RFID wireless sensor both product and resource overuse.
networks and data analytics. The biggest challenge for wide-scale Between now 2050 and beyond, global value chains are facing a
adoption of IoT and associated smart data analytics solutions for the number of major challenges which need to be addressed to ensure a
global food cold chain is the cost of implementation (Angeles, 2005; sustainable and secure supply of food for the world population. As
Jedermann, Ruiz-Garcia, & Lang, 2009; Kelepouris, Pramatari, & previously mentioned, to achieve global food security, it is most es-
Doukidis, 2007). For example, the most important segment, albeit the sential that one adopt a strategy of resource efficiency. Resource effi-
weakest link of the food cold chain is transportation from field to the ciency aims to ensure that production is sustainable across the entire
warehouses and stores. It is well-known that the temperature dis- value chain. It provides an ability to monitor, evaluate and correct any
tribution inside a shipping container is anything but uniform, which value loss streams while at the same time leveraging the value adding
means perishable products placed at different locations inside the for all relevant stakeholders.
container will ultimately have different qualities, nutritional values,
and even bacterial concentrations (James, James, & Evans, 2006). Acknowledgements
Hence, in order to truly benefit from an IoT/big data solution, higher-
resolution monitoring of the shipping container is necessary which re- This publication has emanated from research supported in part by a
quires more sensors and a higher cost of implementation which is an research grant from Science Foundation Ireland (SFI) under Grant


U. Mc Carthy et al. 7UHQGVLQ)RRG6FLHQFH 7HFKQRORJ\²

Number 16/IFB/4439. Gwanpua, S. G., Verboven, P., Leducq, D., Brown, T., Verlinden, B. E., Bekele, E., et al.
This project has received funding from the European Union's (2015). The FRISBEE tool, a software for optimising the trade-off between food
quality, energy use, and global warming impact of cold chains. Journal of Food
Horizon 2020 research and innovation programme under the Marie Engineering, 148, 2–12.
Sklodowska-Curie grant agreement No 70837. Hashemi Beni, L., Villeneuve, S., Leblanc, D. I., Côté, K., Fazil, A., et al. (2012). Spatio-
temporal assessment of food safety risks in Canadian food distribution systems using
GIS. Spatial and Spatio-temporal Epidemiology, 3, 215–223.
References Heap, R. D. (2006). Cold chain performance issues now and in the future. Innovative equip-
ment and systems for comfort and food preservation. Auckland, new-zealand.
Abad, E., Palacio, F., Nuin, M., Zárate, A. G. D., Juarros, A., Gómez, J. M., et al. (2009). Hertog, M. L. A. T. M., Uysal, I., Mc Carthy, U., Verlinden, B. M., & Nicolai Bart, M.
RFID smart tag for traceability and cold chain monitoring of foods: Demonstration in (2014). Shelf life modelling for first-expired-first-out warehouse management. The
an intercontinental fresh fish logistic chain. Journal of Food Engineering, 93, 394–399. Royal Society, 372.
Aborisade, B., & Bach, C. (2014). Accessing the pillars of sustainable food security. Hinton, G. E., Osindero, S., & Teh, Y.-W. (2006). A fast learning algorithm for deep belief
European International Journal of Science and Technology, 3. nets. Neural Computation, 18, 1527–1554.
Adrian, A. M., Norwood, S. H., & Mask, P. L. (2005). Producers' perceptions and attitudes IIR (2009). The role of refrigeration in worldwide nutrition. 5th Informatory Note on
toward precision agriculture technologies. Computers and Electronics in Agriculture, 48, Refrigeration and Food. Available: http://www.iifiir.org/userfiles/file/publications/
256–271. notes/NoteFood_05_EN.pdf [Accessed 26 aUGUST 2016] .
Aghdam, M. S., & Bodbodak, S. (2014). Postharvest heat treatment for mitigation of James, S. J., James, C., & Evans, J. A. (2006). Modelling of food transportation systems –
chilling injury in fruits and vegetables. Food and Bioprocess Technology, 7, 37–53. a review. International Journal of Refrigeration, 29, 947–957.
Angeles, R. (2005). Rfid technologies: Supply-chain applications and implementation Jedermann, R., Behrens, C., Westphal, D., & Lang, W. (2006). Applying autonomous
issues. Information Systems Management, 22, 51–65. sensor systems in logistics—combining sensor networks, RFIDs and software agents.
Asian Development Bank (2013). Food security in Asia and the Pacific. Philippines: Asian Sensors and Actuators A: Physical, 132, 370–375.
Development Bank. Jedermann, R., Nicometo, M., Uysal, I., & Lang, W. (2014a). Reducing food losses by
Badia-Melis, R., Brecht, E., Lowe, A., & Uysal, I. (2013 July). Pallet wide temperature intelligent food logistics. Philosophical Transactions of the Royal Society of London A:
estimation and mapping for perishable food. Poster session presentation at the annual Mathematical, Physical and Engineering Sciences, 372, 20130302.
international meeting of American Society of Agricultural and Biological Engineers Jedermann, R., Palafox-Albarrán, J., Robla, J. I., Barreiro, P., Ruiz-García, L., & Lang, W.
(ASABE), Kansas City, Missouri. (2011). Interpolation of spatial temperature profiles by sensor networks. Sensors,
Badia-Melis, R., Mc Carthy, U., & Uysal, I. (2016). Data estimation methods for predicting 778–781 IEEE, 2011. IEEE.
temperatures of fruit in refrigerated containers. Biosystems Engineering, 151, 261–272. Jedermann, R., Pötsch, T., & Lloyd, C. (2014b). Communication techniques and chal-
http://dx.doi.org/10.1016/j.biosystemseng.2016.09.009. lenges for wireless food quality monitoring. Philosophical Transactions of the Royal
Badia-Melis, R., Ruiz-Garcia, L., Garcia-Hierro, J., & Villalba, J. I. R. (2015). Refrigerated Society of London A: Mathematical, Physical and Engineering Sciences, 372, 20130304.
fruit storage monitoring combining two different wireless sensing technologies: RFID Jedermann, R., Ruiz-Garcia, L., & Lang, W. (2009). Spatial temperature profiling by semi-
and WSN. Sensors, 15, 4781–4795. passive RFID loggers for perishable food transportation. Computers and Electronics in
Badia-Melis, R. J. G. L. R.-G. T. J.-A., Villalba, J. I. R., & Barreiro, P. (2014). Assessing the Agriculture, 65, 145–154.
dynamic behavior of WSN motes and RFID semi-passive tags for temperature mon- Kelepouris, T., Pramatari, K., & Doukidis, G. (2007). RFID-enabled traceability in the food
itoring. Computers and Electronics in Agriculture, 103, 11–16. supply chain. Industrial Management & Data Systems, 107, 183–200.
Carbonneau, R., Laframboise, K., & Vahidov, R. (2008). Application of machine learning Koutsoumanis, K., Pavlis, A., Nychas, G. J., & Xanthiakos, K. (2010). Probabilistic model
techniques for supply chain demand forecasting. European Journal of Operational for Listeria monocytogenes growth during distribution, retail storage, and domestic
Research, 184, 1140–1154. storage of pasteurized milk. Applied and Environmental Microbiology, 76, 2181–2191.
Carullo, A., Corbellini, S., Parvis, M., & Vallan, A. (2009). A wireless sensor network for LeBlanc, D. I., Villeneuve, S., Hashemi Beni, L., Otten, A., Fazil, A., Mckellar, R., et al.
cold-chain monitoring. IEEE Transactions on Instrumentation and Measurement, 58, (2015). A national produce supply chain database for food safety risk analysis.
1405–1411. Journal of Food Engineering, 147, 24–38.
Coates, A., Lee, H., & Ng, A. Y. (2011). An analysis of single-layer networks in un- Lee, H., Grosse, R., Ranganath, R., & Ng, A. Y. (2009). Convolutional deep belief networks
supervised feature learning. 14 th international conference on artificial intelligence and for scalable unsupervised learning of hierarchical representations. Proceedings of the
statistics (AISTATS) Fort Lauderdale, FL, US. 26th annual international conference on machine learning. Montreal, Quebec, Canada:
Cortes, C., & Vapnik, V. (1995). Support-vector networks. Machine Learning, 20, 273–297. ACM.
Dabbene, F., Gay, P., & Tortia, C. (2014). Traceability issues in food supply chain man- Li, S., Xu, L. D., & Zhao, S. (2015). The internet of things: A survey. Information Systems
agement: A review. Biosystems Engineering, 120, 65–80. Frontiers, 17, 243–259.
Dargie, W., & Poellabauer, C. (2010). Fundamentals of wireless sensor Networks: Theory and Maggio, A., Criekinge, T. V., & Malingreau, J. P. (2015). In J. S. A. P. Reports (Ed.). Global
practice. John Wiley & Sons. Food Security 2030-Assessing trends with a view to guiding future EU policies.
DeNavas-Walt, C., & Proctor, B. D. (2014). Income and poverty in the United States: 2013 Mahajan, P. V., Caleb, O., Singh, Z., Watkins, C., & Geyer, M. (2014). Postharvest treat-
current population reports. ments of fresh produce. Philosophical Transactions of the Royal Society of London A:
Deng, L., & Yu, D. (2014). Deep Learning: Methods and applications. Now Publishers. Mathematical, Physical and Engineering Sciences, 372, 20130309.
Derens, E., Palagos, B., & Guilpart, J. (2006). The cold chain of chilled products under Martincus, C. V., & Graziano, A. (2012). Customs as Doorkeepers: What are their effects on
supervision in France. IUFOST 13th world congress of food science & Technology: Food is international trade? Inter - American Development Bank.
life (pp. 51–64). France: Nantes. Mc Carthy, U., Ayalew, G., Corkery, G., Laniel, M., Uysal, I., Butler, F., et al. (2012). RFID
Dou, Z., Ferguson, J. D., Galligan, D. T., Kelly, A. M., Finn, S. M., & Giegengack, R. (2016). and the EPCGlobal network as a pivotal tool in the development of a sustainable
Assessing U.S. food wastage and opportunities for reduction. Global Food Security, 8, organisational compeditive advantage. In W. D. Nelson (Ed.). Advances in business and
19–26. management. Nova Publishers.
Dudani, S. A. (1976). The distance-weighted k-nearest-neighbor rule. IEEE Transactions on McCarthy, U., Ayalew, G., Butler, F., Mcdonnell, K., & Ward, S. (2011). The case for UHF
Systems, Man, and Cybernetics, 325–327 SMC-6. RFID application in the meat supply chain in the Irish context: A review perspective.
ERS, U. (2017). International food security assessment, 2017-2027. Agricultural Engineering International: the CIGR Ejournal, 13.
EU (2015). In E. Comission (Ed.). Closing the loop - an EU action plan for the Circular McCarthy, U., Uysal, I., Laniel, M., Corkery, G., Butler, F., Mcdonnell, K., et al. (2013).
Economy. Brussels: European Comission. Sustainable global food supply networks. In T. Norton, B. Tiwari, & N. Holden (Eds.).
EU (2016). Food and drink industry [Online]. Available: http://ec.europa.eu/growth/ Sustainable food processing. Wiley-Blackwell.
sectors/food/ [Accessed 28 August 2016 2016] . McClelland, J. L., & Rumelhart, D. E. (1988). Explorations in parallel distributed processing:
FAO (2011). Global food losses and food waste – extent, causes and prevention. A handbook of models, programs, and exercises.
FAO (2014). Family Farmers Feeding the world, caring for the earth. McKellar, R. C., Leblanc, D. I., Rodríguez, F. P., & Delaquis, P. (2014). Comparative si-
FAO (2015a). In F. A. A. O. O. T. U. Nations (Ed.). FAO statistical pocketbook 2015 ed. mulation of Escherichia coli O157:H7 behaviour in packaged fresh-cut lettuce dis-
FAO (2015b). The state of food and agriculture. Social protection and agriculture: Breaking the tributed in a typical Canadian supply chain in the summer and winter. Food Control,
cycle of rural poverty. Food And Agriculture Organization Of The United Nations. 35, 192–199.
FAO (2015c). The state of food insecurity in the world 2015. Meeting the 2015 international Mcmurray, G., Arruda, C., Britton, D., Eidenberger, T., Evans, S., Gibney, M., et al. (2013).
hunger targets: Taking stock of uneven progress. Rome. I., WFP. Food security: A systems approach. EU science. Global Challenges and Global
Foster, A. M., Swain, M. J., Barrett, R., & James, S. J. (2003). Experimental verification of Collaboration.
analytical and CFD predictions of infiltration through cold store entrances. Mercier, S., Villeneuve, S., Mondor, M., & Uysal, I. (2017). Time–temperature manage-
International Journal of Refrigeration, 26, 918–925. ment along the food cold chain: A review of recent developments. Comprehensive
Friedman, N., Geiger, D., & Goldszmidt, M. (1997). Bayesian network classifiers. Machine Reviews in Food Science and Food Safety, 16, 647–667.
Learning, 29, 131–163. Michalski, R. S. (1983). A theory of Methodology of inductive learning,” machine Learning: An
Giannakourou, M. C., & Taoukis, P. S. (2003). Application of a TTI-based distribution artificial intelligence approach. Springer-Verlag Berlin.
management system for quality optimization of frozen vegetables at the consumer Mohri, M., Rostamizadeh, A., & Talwalkar, A. (2013). Foundations of machine learning. The
end. Journal of Food Science, 68. MIT Press.
Gubbi, J., Buyya, R., Marusic, S., & Palaniswami, M. (2013). Internet of things (IoT): A do Nascimento Nunes, M. C., Nicometo, M., Emond, J. P., Melis, R. B., & Uysal, I. (2014).
vision, architectural elements, and future directions. Future Generation Computer Improvement in fresh fruit and vegetable logistics quality: Berry logistics field stu-
Systems, 29, 1645–1660. dies. Philosophical Transactions of the Royal Society A: Mathematical, Physical &
Gunders, D. (2012). Wasted: How America is losing up to 40 percent of its food from farm to Engineering Sciences, 372.
fork to landfill. Natural Resources Defense Council. NRDC (2016). Natural resources defense council. [Online] . Available: http://www.


U. Mc Carthy et al. 7UHQGVLQ)RRG6FLHQFH 7HFKQRORJ\²

savethefood.com/http://www.savethefood.com/ [Accessed 28 August 2016 2016] . gov/pmc/articles/PMC2430252/ [Accessed 28 August 2016] .

OECD (2013a). Aid for trade and value chains in transport and logistics. Tromp, S.-O., Rijgersberg, H., Pereira da Silva, F., & Bartels, P. (2012). Retail benefits of
OECD (2013b). Interconnected Economies: Benefiting from global value chains – synthesis dynamic expiry dates—simulating opportunity losses due to product loss, discount
report. policy and out of stock. International Journal of Production Economics, 139, 14–21.
Olsen, P., & Borit, M. (2013). How to define traceability. Trends in Food Science & USDA (2016). Ag and food sectors and the economy [Online]. Available: http://www.ers.
Technology, 29, 142–150. usda.gov/data-products/ag-and-food-statistics-charting-the-essentials/ag-and-food-
Petersen, A., & Green, D. (2005). Seafood traceability: A practical guide for the US industry. sectors-and-the-economy.aspx [Accessed 22 August 2016 .
North Carolina: National Fisheries Institute, Inc and North Carolina Sea Grant. Van Dorp, K.-J. (2002). Tracking and tracing: A structure for development and con-
Piramuthu, S. (2005). Knowledge-based framework for automated dynamic supply chain temporary practices. Logistics Information Management, 15, 24–33.
configuration. European Journal of Operational Research, 165, 219–230. Vapnik, V. N. (1998). Statistical learning theory.
Qi, L., Xu, M., Fu, Z., Mira, T., & Zhang, X. (2014). C 2 SLDS: A WSN-based perishable Vared Sarah, F. E. (2016). A roadmap to reduce u.s. Food waste by 20 percent.
food shelf-life prediction and LSFO strategy decision support system in cold chain Weiser, M. (1999). The computer for the 21st century. Sigmobile Mob. Computer
logistics. Food Control, 38, 19–29. Communication Review, 3, 3–11.
Rabiner, L. R. (1989). A tutorial on hidden Markov models and selected applications in Welbourne, E., Battle, L., Cole, G., Gould, K., Rector, K., Raymer, S., et al. (2009).
speech recognition. Proceedings of the IEEE, 77, 257–286. Building the internet of things using RFID: The RFID ecosystem experience. IEEE
Regattieri, A., Gamberi, M., & Manzini, R. (2007). Traceability of food products: General Internet Computing, 13, 48–55.
framework and experimental evidence. Journal of Food Engineering, 81, 347–356. World-Bank (2012). In A. Jean-Francois, M. M. Alina, O. Lauri, S. Ben, & S. Daniel (Eds.).
Salakhutdinov, R., & Hinton, G. (2009). In T. I. C. O. A. I. A. S. Aistats (Ed.). Deep Connecting to compete 2012-trade logistics in the global economy. Washington, DC
Boltzmann machines Clearwater Beach, Florida, USA. 20433: The International Bank for Reconstruction and Development/The World
Sánchez-López, T., & Kim, D. (2008). Wireless sensor networks and RFID integration for Bank.
context aware services. White paper series 2008; AUTOIDLABS-WP-SWNET-026, Worldfoodday (2016). Ending hunger starts with us [Online]. Available: http://www.
Daejeon, Korea 2008. worldfooddayusa.org/what-is-wfd [Accessed 28 August 2016 2016] .
Spink, J., Moyer, D. C., & Speier-Pero, C. (2016). Introducing the food fraud initial WRI (2013). In W. R. Institute (Ed.). Creating a Sustainable Food Future - a menu of solutions
screening model (FFIS). Food Control, 69, 306–314. to sustainably feed more than 9 billion people by 2050. World Resource Institute.
Stephens, E. C., Jones, A. D., & Parsons, D. (2018). Agricultural systems research and Young, L. (2012). Our biggest problem? We’re wasting food. Canada tosses 40 per cent of the
global food security in the 21st century: An overview and roadmap for future op- food we produce each year. It's bad for business and bad for the planet. November 2012.
portunities. Agricultural Systems, 163, 1–6. http://dx.doi.org/10.1016/j.agsy.2017. Available: http://www.canadiangrocer.com/top-stories/what-a-waste-19736
01.011. [Accessed 26 August 2016] .
Tenenbaum, D. J. (2008). Food vs. Fuel: Diversion of crops could cause more hunger. Zhou, Z.-H., Wu, J., & Tang, W. (2002). Ensembling neural networks: Many could be
Environmental health perspectives. [Online], 116. Available: http://www.ncbi.nlm.nih. better than all. Artificial Intelligence, 137, 239–263.