Frontiers in Business, Economics and Management 
ISSN: 2766-824X | Vol. 4, No. 2, 2022 

 

91 

Evaluation	and	Analysis	of	Food	System	Based	on	ESE	
Model	
Yanxia Wang1, a 

1Department of Electrical Engineering, North China Electric Power University, Baoding 071000, China 
a1994505926@qq.com 

 

Abstract: In order to better analyze the food system and alleviate hunger, we mainly design an ESE model. First, we select 16 
indicators from three fields: economic, social and environment. After data processing and weight calculation, we obtained long-
term and short-term scores of a food system. We analyze three typical food systems in the world today and select a countries for 
analysis. Finally, we analyze the sensitivity of the intervention model. 

Keywords: Weight Calculation Method, Intervention Model, Sensitivity. 
 

1. Introduction 
Zero Hunger, goal 2 of the SDGs, promises to end hunger, 

achieve food security, improve nutrition and promote 
sustainable agriculture, which is also the primary work of the 
WFP. 

However, the current food system, which allows food to be 
produced and distributed relatively cheaply and efficiently, 
prioritizes efficiency and profitability. Although the system is 
highly efficient, it has resulted in serious inequities and a 
residual environmental footprint. 

To better achieve the goal of zero hunger, and to address 
the inequity and unsustainability of the current food system, 
it is imperative to reimagine and reorder the priorities of it. 
Therefore, we design an ESE model to solve this problem. 

2. Economics- Social- Environmental 
Model 

In assessment of the food system of in a country, we need 
to consider many factors to comprehensively reflect the 
efficiency, profitability, sustainability and equity of it. 

In this paper,We have determined sixteen main factors. 
These factors can be divided into three main fields: 
economics, social and en-vironmental. We will first explain 
how to quantify the influence of each factor, then describe 
how to determine the weight of each indicator and how to 
process data, and finally calculate the   in different countries. 

2.1. Economics 
Here are six economic indicators that directly measure the 

level of food production and consumption in a country. And 
then, they can represent the efficiency and profitability of the 
food system. We only consider the economic indicators most 
closely related to the food system and ignore other minor 
factors. 

2.1.1. Food Available Per Capita 
The amount of food available per capita is the foundation 

of a robust food system. Only do ensure the amount of food 
available per capita can further discuss the food consumption 
capacity. Therefore we can choose the food production index, 

( ), to represent the amount of food available per capita. 

2.1.2. Food Expenditure 
Food expenditure is a direct reflection of a country's ability 

to consume food. Food consumption is basic and does not 
augment with an increase in income or consumption. Engel's 
coefficient refers to the proportion of food expenditure in total 
living expenses in a household, which decreases with the 
increase of family income. 

Since it is not easy to obtain the data of food expenditure, 
we use the product of Engel's coefficient and the total living 
expenditure of residents to quantify the food expenditure 

index( ). The calculation is shown in Eq.1. 

 
  (1) 

 

2.1.3. Food Prices 
Like food expenditure, food prices directly show the food 

consumption capacity of a country, too. Therefore, we choose 

the food price index( ), , to represent the food prices. 

2.1.4. Food Diversity 
The main types of food vary from country to country. We 

set the food consumption diversity degree(  ),  , to 

quantify the level of food consumption structure. The 
calculation formula can be seen in Eq.2. 

 

       (2) 

 

Where,   is proportion of meat, fish and aquatic 

products,   is proportion of eggs and egg products, 

 is proportion of dairy and dairy products,  is 

proportion of food crops like wheat, rice and corn, 

 is proportion of vegetables and fruits. 

2.1.5. Food Sources 
Food sources include both domestic production and import. 

We use the proportion of imported food to total food( ), , 

to indicate the degree of diversity of food sources. 

2.1.6. Agricultural Development 
Agriculture is the basic industry that provides support for 



 

92 

the construction and development of national economy, and 
agricultural development is the foundation of a country's food 

system. We choose the agricultural mechanization level( ), 

, to quantify the agricultural development of a country. 

The calculation expression for economics index of our ESE 
model is： 

 
 (3) 

 

2.2. Social 
We use six social indicators to quantify the impact of 

society on food system. These social indicators are required 
to reflect the equity of the country in terms of food production, 
distribution and consumption. 

2.2.1. Vulnerable Groups Food Security 
The food security of vulnerable groups is an important 

factor to measure the equity of the food system. The more 
equal the food system is, the more guaranteed the food of 
vulnerable groups will be. When people have less real income, 
they spend less on food and adjust their consumption 
accordingly. The effect gets stronger with less income, which 
means life gets harder for the vulnerable. Clearly, the food 
security of vulnerable groups is directly related to the 

vulnerable employment rate( ) . In addition, in order to 

quantify the food security of vulnerable groups, the 

government subsidies, ( ), should also be considered. 

2.2.2. Food Nutrition 
Food insecure means being without reliable access to 

sufficient affordable and nutritious food. We can process the 
data related to food insecurity to measure the food nutrition 
of the country. At the same time, the United Nations focuses 
on the welfare of children. So, we also take the impact of the 
proportion of healthy children under five into account. Eq.4 

is used to quantify residents' food nutrition level( ), . 

 
 (4) 

 

Where,   is the prevalence of severe food 

insecurity in the population (%),  

is the prevalence of undernourishment (% of children under 

five), and  is the prevalence of overweight, (% 

of children under five) 

2.2.3. Food Policy 
Compared with higher-income countries, lower middle-

income countries are less suited to national food policies, 
resulting in huge differences in the structure of food systems. 
We will combine the data of CPIA policies for social 
inclusion/equity cluster average retrieved from the World 
Bank database with the actual situation of national food 

policies, and then give a score( ), , from 1 (lowest) to 6 

(highest). 

2.2.4. Conflict( ) 

More than half of the hungry people around the world live 
in countries affected by conflict. If poverty, unemployment or 
economic hardship strike at the same time, hunger can 
exacerbate the conflict. The vicious cycle of conflict and 

hunger must be broken if we are to create a world of food 
security for all,as Eq.5. 

 

      (5) 

 

Where   is the battle-related death toll, 

  is the displaced persons by conflict and 

violence,   is refugee population by country or 

territory of asylum. 

2.2.5. Income Gap 
The world produces enough food to feed everyone, 

however, many people in the world still do not have enough 
income to buy enough food or to obtain nutritious food. 

Therefore ，  We use the Gini coefficient(  ),  , to 

represent the income gap, and then show the food inequality. 
The mathematical expression for social indicators in the 

ESE model equation has a form of: 
 

   (6) 
 

2.3. Environment 
We assign four environmental indicators. These widely 

used indicators allow us to capture key aspects of the national 
environmental dimension in food system, which further 
allows us to assess the sustainability of food system. 

2.3.1. Food Wastage 
Curbing food wastage is critical to sustainability of food 

system. We use the food wastage degree(  ),  , to 

represent this indicator. 

2.3.2. Food Recovery 
Food recovery rate(  ),  , refers to the proportion of 

food actually recycled to the total amount of food wastage. 
Developing the recycling of wasted food can reduce the 
impact of food wastage on the environment to a certain extent, 
thus drawing near to sustainable development.  

2.3.3. Environmental Footprint 
It is important to reduce the depletion of natural resources 

in the production and consumption of food. We quantify the 
environmental footprint(  ),   from four aspects: fresh 

water, soil, greenhouse gas emissions, and biodiversity. Its 
calculation formula is as follows: 

 
  (7) 

 

In Eq.7,   is the proportion of agriculture 

freshwater to total freshwater withdrawal,   is the 

proportion of agricultural land area to total land area,  

is the proportion of greenhouse gas emissions caused by food 

system, and  is the proportion of biodiversity 

loss caused by food system, respectively. 

2.3.4. Food Saving Awareness 
Food saving awareness plays an important role in the 

sustainable development of food systems. The general food 

saving awareness(  ),  ,of citizens may be affected by 

their degree of education, government advocacy, efforts made 



 

93 

by social welfare organizations and so on. In view of the 
uncertainty of other factors, we will only focus on the first 
two factors. They are the average degree of education of 

citizens,  , and the government advocacy of 

food saving, . 

 

    (8) 

 
The mathematical expression for environmental indicators 

in the ESE model equation has a form of: 
 

 (9) 

 

2.4. Data Processing 
2.4.1. Data Acquisition and Preprocessing 

We searched databases like World Bank, United Nations 
World Food Programme and so on [11-12]. And we found basic 
data for 16 indicators across 126 countries 

However, not all data can be collected and statistics for 
some indicators are lost. In order to improve this situation, we 
use SPSS for data preprocessing. 

2.4.2. Data Normalization 
In the process of analyzing the indicators, we find that all 

the indicators can be divided into three categories. Tag ‘ ’ 

means that for the indicator, bigger is better. Similarly, tag 

‘  ’ means smaller is better, and tag ‘  ’ means that the 

value is better when it is closer to the specific value. 
At the same time, all indicators are different in dimension, 

so they cannot be directly compared. For the convenience of 
subsequent calculations, we need to standardize these sixteen 
indicators, converting them into numbers in the range of 0-1. 

We will deal with economic, social and environmental 
indicators as follows: 

Let  denotes the j-th indicator of the i-th country. 

As for indicators those smaller is better tagged as , the 

equation (10) should be: 
 

       (10) 

 
As for indicators those closer to the specific value is better 

tagged as , the equation (11) should be: 

 

   (11) 

 
Where  is the best value of the j-th indicator. 

After unifying all indicators into indexes that bigger is 
better, we use the following Eq.12 to standardize the indexes, 
eliminate the dimensionality effect, and convert them into 
numbers within the range of 0-1. 

 

              (12) 

 

2.4.3. Weights Calculation 
Different indicators have different effects on the food 

system of a country. We look for index data in the fields of 
economics, society and environment, use the entropy weight 
method (EWM) to calculate the weights of massive data, and 
then apply the index after weight calculation to Eq.3, Eq.6 and 
Eq.9. 

EWM is an objective weighting method. The smaller the 
variation degree of the index, the less information it reflects, 
and the lower the corresponding weight. 

For the normalized matrix   that has been processed 

from the three fields of economics, society and environment, 

we will calculate their weights, , respectively. 

First, we calculate the probability matrix  , where the 

Eq.15 for each element  in  is as Eg.13. 

 

            (13) 

 
Then, we calculate the information entropy of j-th indicator, 

as Eq(14). 
 

        (14) 

 
Finally, we can get the entropy weight of the j-th 

indicator,as Eg(15). 
 

            (15) 

 

2.5. ESE Score 
By calculating the scores and weights of different 

indicators, we have obtained the scores of economic, social 
and environmental fields. 

We define the scoring formula of ESE as Eq(16): 
 

 (16) 
 
Through different weight calculation methods, we can get 

different values of ,  and . So, we can evaluate 

the costs and benefits of a food system from different aspects.  

3. Impact of Changing Priorities 
Different countries have different conditions, so they have 

different needs for their food systems. We will analyze the 
necessity to change food system priorities in terms of short-
term benefits and long-term costs in the light of practical 
development needs. 



 

94 

3.1. Short-term Impact 
In order to better compare the current situation of the food 

system, we adopt the idea of weighted average method, as 
shown in Eq.17. And then we can figure out the total score of 
food system in the short run, . 

 

      (17) 

 
The weighted average method amplifies the impact of 

short-term benefits.   calculated after 

adopting the weighted average can reflect the appearance of 
the food system in the short term, that is, short-term benefits. 
Therefore, it is easier to see the superiority of the current 
system priority selection. 

3.2. Long-term Impact 
In our view, the impacts of economics, society and 

environment on the food system should trend equally in the 
long run. So, we define: 

 

         (18) 

 
Then we can figure out the total score of food system in the 

long run, . 

By looking at the  in long-term trends, we 

can see why countries were intervening, and we can also 
make recommendations for intervention in countries that are 
not. 

3.3. Typical Cases Analysis 
We evaluate food systems in terms of both short-term 

benefits and long-term costs. According to the above ESE 
model, we calculated the long-term and short-term ESE 
scores of different countries, that is   and 

 . And we also predict the general trend 

according to its scores of  ,   and 

. 
Our model shows a wide range of national food systems, 

but the overall trend is in three directions. These three typical 
food systems are selected for our analysis, that is, the typical 
developed country, the typical developing country in conflict 
and the typical developing country in stability, as shown in 
Figure 1. 

 

 
(a) developed country 

 
(b) developing country in conflict 

 

 
(c) developing country in stability 

Figure 1. ESE Scores 
 
Case: Developing Countries in Conflict------ Afghanistan 
Some extremely hungry developing countries live in areas 

of perpetual conflict, where conflict and political instability 
exacerbate food system instability. In severe areas, the food 
system is on the verge of collapse due to the conflict, and there 
is a serious bipolar phenomenon. Compared with other 
countries, the immediate priority for developing countries in 
conflict is to solve the problem of hunger and help the 
vulnerable. 

In Afghanistan, for example, with the war, conflict and a 
healthy diet hidden costs increase, the number of hungry 
people is on the increase in Afghanistan. As shown in Fig. 2, 

  and   of the Afghan food 

system are both low. It is clear that, the Afghan food system 
is so unstable because of the severe conflict and inequity. It 
may even be on the verge of collapse in the long run. So, with 
the aim of alleviating the hunger crisis, it is crucial to optimize 
the food system of Afghanistan. 

 



 

95 

 
Figure 2. ESE Scores in Afghanistan without intervention 

 
Based on the actual situation, we present the most effective 

optimization scheme for improving the food system in 
Afghanistan. 

3.3.1. Aid Interventions 
Conflicts and wars have weakened the handling ability of 

government. Therefore, assistance programs from the UN and 
international organizations must be used to help alleviate the 
hunger crisis in Afghanistan. 

We quantify the level of aid intensity,  , which 

reduces the gap between rich and poor, to demonstrate the 
effectiveness of optimizing food systems. We use the initial 

value  for the simulation, and we believe that the 

level of aid intensity will increase to alleviate the hunger crisis. 
Clearly, social influence decreases over time. Since aid 
intervention will directly lead to the deceleration of the 
growth trend of the gap between the rich and the poor, that is, 

the growth trend of social influence curve  slows down. 

We set the following relationship: 
 

    (19) 

 

Here we define .  

3.3.2. Policy Interventions 
As an agricultural country, agriculture contributes more 

than one third of Afghanistan economic output. So, making 
food systems more sustainable is essential. The government 
should take measures to promote the sustainable development 
of the food system. 

We use  to simulate the process of increasing policy 

intensity, which is similar to that of aid interventions. We set 
the intervention as follows: 

 

    (20) 

 

Here we define . 

After integrating the needs of the country to optimize the 
food system, we get the simulation curve fitting results as 
Figure 3. Obviously, measures to intervene in the food system 
of Afghanistan are necessary. 

 

 
Figure 3.  in Afghanistan 

4. Sensitivity Analysis 

When determining the intervention intensity parameter , 

we referred to online databases and literature, and combined 
with the actual situation of the country. In this section, we test 
the sensitivity of the model by changing the value of 

parameter , to demonstrate its robustness. 

 



 

96 

 
Figure 4. Sensitivity analysis 

 
When analyzing Afghanistan, we set parameter 

, to show the intensity of the intervention. In Fig. 

4, we change the value of  from 0.2 to 0.4, and find that 

the trend does not change significantly. So, we think our 
model is reliable. 

5. Conclusion 

5.1. Strengths and Weaknesses 
1.When building the model, we used SPSS and MATLAB 

to preprocess and standardize massive data. Through a 
complex but meaningful set of pretreatments, the benefits of 
the food system in the economics, society, and environment 
could be reflected. 

2.Our model has some flexibility. Through different weight 
calculation methods in Eq.16, we can evaluate the advantages 
and disadvantages of a food system from different aspects. 

3.Our intervention model passed the sensitivity analysis 
and was robust. So, the implementation time of an optimal 
food system can be predicted for each country. 

4.Our model cannot account for the effect of Covid-19. As 
it is still difficult to fully under-stand the potential damage 
caused by the Covid-19 pandemic, it should be recognized 
that  

there are still significant uncertainties in the current 
assessment projections.  

5.We didn't consider vegetarians, people with allergies, or 
individual differences such as gender and race. This can be 
improved by adding the regulatory factor of individual 
difference. Similarly, we don't consider the processing and 
packaging of the food. This can be improved by adding the 
regulatory factor of time lag to some relevant indicators. 

5.2. Model Development 
Due to the flexibility of indicators selection and weights 

calculation methods, our mod-el has certain generalizability. 
Our model can be applied to smaller food systems, broader 
regions, and other biologically related mathematical problems. 

Through this intervention model, we can simulate the 
growth curve of population and the intervention effect of 
corresponding policies on population. 

References 
[1] FAO, IFAD, UNICEF, WFP, WHO. In 2020. The State of 

Food Security and Nutrition in the World by 2020: 
Transforming the Food System and Securing Healthy, 
Economic Meals - Summary. Rome. 

[2] ZHANG J Q. The status of the food system in Japan and its 
research progress. Journal of Food Science and Technology, 
2007, 26(2): 115-120. 

[3] GUO H, WANG L E, MA E P. The evolving concept of food 
system and a geographical paradigm of research[J]. Progress in 
Geography, 2019,38(07):1034-1044. 

[4] FAN S G, GAO H X. Rethinking the global food system under 
covid-19[J]. Journal of Huangzhong Agricultural University 
(Social Science Edition) ,2020(05):1-8+168. 

[5] LIU J J, LI B F, NING Y F, ZHANG W N. The concept, theory, 
and strategies of food ur-banism[J]. Planners, 2012,28(03):91-
95. 

[6] Databank. Retrieved from: 
https://databank.shihang.org/home.aspx 

[7] World Food Programme (WFP). Retrieved from: 
http://www.wfp.org/