Professional Documents
Culture Documents
Sustainability:
During the 60’s it was expected that in the near future, there would not be
enough food to cope with the growing population. During the late 70’s, global
food production grew to a level that was able to cope with the increasing
population. Since then, the focus has been shifted towards the problems of
insufficient food supplies in major parts of the world and concerns over intensive
farming. Due to this, there has been an increasing amount of interest into long
term food security due to problems such as land degradation.
As a result of the need to target levels of pollution and sustainable farming, the
UN introduced a programme titled Agenda 21. This however, was seen as too
general for local scales and lead to the Brundtland report. This ensured that the
later generations would be able to meet there own farming needs whilst still
working towards improvement. The Food and Agricultural Organisation (FAO) of
the UN have also offered a more specific description to lead towards sustainable
farming that is more socially acceptable.
In various practices the Total Factor Productivity (TFP) is used as a yardstick for
comparison; however there are others.
As the quality of the soil is one of the main contributing factors to the yield of the
crop, it is important at maintain it at a reasonable level. The condition of the soil
is improved by actively coping with soil pollution through regulatory and market
measures. It is also improved by seeking feasible solutions for clean up areas
e.g. implementing the bronwfield policy to repair damaged areas..
Since rough set theory is in the relatively early stages, there are some problems:
In this study, rough set theory was used to determine whether there are factors
that systematically affect the variation in productivity in agricultural and agro-
food sector.
Application:
Two tables are shown in the document listing the productivity growth in the farm
sector for OECD’s, and the growth in the productivity growth in the agro-food
sector. The data from these tables then undergoes a rough set analysis. This is
undertaken by classifying the codification for attributes and decision variables.
The codes for these variables are shown in tables 3 and 4. Tables 5 and 6
present the results.
Table 1 Studies of productivity growth in the farm sector of the OECD countries
Outpu
N Country Input t Time Period TFP
1 Australia -0.10 1.10 1971 1981 1.20
2 Australia -0.10 1.90 1971 1990 2.00
3 Canada -0.10 2.50 1981 1990 2.60
4 Canada -1.06 1.17 1962 1970 2.23
5 Canada 0.01 2.37 1962 1978 2.36
6 Canada 1.05 3.57 1970 1978 2.52
7 Canada -0.40 2.60 1962 1971 3.00
8 Canada 2.40 0.20 1962 1990 -2.20
9 Canada 1.40 2.60 1971 1981 1.20
1
0 Canada 1.90 -0.50 1981 1990 -2.40
1
1 Switzerland -1.20 1.10 1973 1981 2.30
1
2 Switzerland -1.10 0.90 1973 1988 2.00
1 Switzerland -1.10 0.00 1981 1988 1.10
3
1
4 U.S. 0.88 1.85 1953 1957 0.97
1
5 U.S. 0.25 2.92 1957 1960 2.67
1
6 U.S. 0.24 1.90 1960 1969 1.66
1
7 U.S. 0.68 1.06 1969 1973 0.38
1
8 U.S. 0.72 2.75 1973 1979 2.03
1
9 U.S. 0.50 2.20 1973 1988 1.70
2
0 U.S. 0.08 1.21 1962 1970 1.13
2
1 U.S. 0.31 1.72 1962 1978 1.41
2
2 U.S. 0.55 2.22 1970 1978 1.67
2
3 U.S. 0.14 1.66 1950 1960 1.52
2
4 U.S. 0.17 1.76 1950 1982 1.59
2
5 U.S. 0.05 0.84 1960 1970 0.79
2
6 U.S. 0.34 2.60 1970 1982 2.26
2
7 Germany -0.30 1.60 1967 1987 1.90
2
8 France 0.10 2.30 1967 1987 2.20
2
9 The Netherlands 1.40 4.00 1967 1987 2.60
3
0 Belgium -0.20 1.80 1967 1987 2.00
3
1 Luxembourg -1.90 1.80 1967 1987 3.70
3
2 U.K. -0.20 1.60 1967 1987 1.80
3
3 Ireland 0.80 2.60 1967 1987 1.80
3
4 Italy 0.80 1.80 1974 1987 1.00
3
5 Denmark 0.70 2.60 1974 1987 1.90
3
6 Germany -0.40 1.30 1973 1989 1.70
3
7 France -0.50 1.90 1973 1989 2.40
3
8 Italy -0.60 1.70 1973 1989 2.30
3
9 The Netherlands 0.40 3.10 1973 1989 2.70
4 Belgium/Luxemb 0.00 1.20 1973 1989 1.20
0 ourg
4
1 U.K. -0.20 1.50 1973 1989 1.70
4
2 Ireland 0.90 2.30 1973 1989 1.40
4
3 Denmark -0.10 2.30 1973 1989 2.40
4
4 Germany 0.00 1.70 1965 1985 1.70
4
5 France 0.10 1.80 1965 1985 1.70
4
6 Italy -0.60 1.60 1965 1985 2.20
4
7 The Netherlands 2.60 4.20 1965 1985 1.60
4 Belgium/Luxemb
8 ourg 0.00 2.20 1965 1985 2.20
4
9 U.K. -0.30 2.20 1965 1985 2.50
5
0 Ireland 0.90 2.90 1965 1985 2.00
5
1 Denmark -0.10 1.50 1965 1985 1.60
5
2 The Netherlands 1.40 3.60 1950 1960 2.20
5
3 The Netherlands 0.10 3.80 1960 1970 3.70
5
4 The Netherlands 1.10 4.40 1970 1980 3.30
5
5 The Netherlands -0.50 2.40 1980 1988 2.90
5
6 The Netherlands 0.60 3.60 1950 1988 3.00
Condition Attributes:
A1 - Country Code
Other Canada 1
Australia 1
USA 1
Large European UK 2
Countries Germany 2
Italy 2
France 2
Small European Denmark 3
Countries Finland 3
Netherlan
ds 3
Sweden 3
Norway 3
Belgium 3
Code
A2 - Less than -0.5 1
Input Between -0.4
and -0.1 2
Between 0.0 and
0.4 3
Between 0.5 and
0.9 4
More than 1.0 5
Code
A3 - Less than 1.5 1
Output Between 1.6
and 2.6 2
More than 2.5 3
Time period:
Code
A4 - Starting Before
Point 1960 1
1970 2
1980 3
Code
A5 - 10
Length Years 1
20
Years 2
30
Years 3
Decision Variables:
Code
Total Factor Less than 1.5 1
Productivity Between 1.6 and
2.5 2
More than 2.5 3
Condition Attributes:
Cod
A1 - Country e
Other Canada 1
Australia 1
USA 1
Asia Japan 2
Large European UK 3
Countries Germany 3
Italy 3
France 3
Small European Denmark 4
Countries Finland 4
Netherla
nds 4
Sweden 4
Norway 4
Belgium 4
Cod
e
A2 - Sector Food 1
Coverage Beverages 2
Manufacturing 3
Food and Beverages 4
Food, Beverages and
Tobacco 5
Agriculture 6
Cod
e
A3 - Method Index number 1
Used Econometric cost function 2
Econometric production
function 3
Input-output 4
Cobb-Douglas production
function 5
Solow growth accounting
method 6
Labour productivity 7
Time Period:
Co
de
A4 - Starting 19
Point 50 1
19
60 2
19 3
70
19
80 4
Cod
e
A5 - 10
Length Years 1
20
Years 2
30
Years 3
Decision Variables:
Cod
e
Average Percentage Less than 1% 1
Change Between 1% and
2% 2
More than 2% 3
Table 5 Information system of the studies related to the farm sector after
codification
Object
s A1 A2 A3 A4 A5 D
1 1 2 1 2 1 1
2 1 2 2 2 2 2
3 1 2 2 3 1 3
4 1 1 1 1 1 2
5 1 3 2 1 2 2
6 1 5 3 2 1 3
7 1 2 3 1 1 3
8 1 5 1 1 3 2
9 1 5 3 2 1 1
10 1 5 1 3 1 2
11 3 1 1 2 1 2
12 3 1 1 2 1 2
13 3 1 1 3 1 2
14 1 4 2 1 1 3
15 1 3 3 1 1 3
16 1 3 2 1 1 3
17 1 4 1 2 1 1
18 1 4 3 2 1 2
19 1 4 2 2 1 2
20 1 3 1 1 1 1
21 1 3 2 1 1 1
22 1 4 2 2 1 2
23 1 3 2 1 1 1
24 1 3 2 1 3 1
25 1 3 1 1 1 1
26 1 3 3 2 1 2
27 2 2 3 1 1 3
28 2 3 2 1 2 2
29 3 5 3 1 2 3
30 3 2 2 1 2 2
31 3 1 2 1 2 3
32 2 2 2 1 2 2
33 3 4 3 1 2 2
34 2 4 2 2 1 3
35 3 4 3 2 1 2
36 2 2 1 2 2 2
37 2 1 2 2 2 2
38 2 1 2 2 2 2
39 3 3 3 2 2 2
40 3 3 1 2 2 1
41 2 2 1 2 2 2
42 3 4 2 2 2 1
43 3 2 2 2 2 2
44 2 2 2 1 1 2
45 2 3 2 1 2 2
46 2 1 2 1 2 2
47 3 5 3 1 2 1
48 3 3 2 1 2 2
49 2 2 2 1 2 2
50 3 4 3 1 2 2
51 3 2 1 1 2 2
52 3 3 3 2 1 2
53 3 3 3 1 1 3
54 3 5 3 2 1 3
55 3 1 2 3 1 3
56 3 4 3 1 3 3
Table 6 Information system of the studies related to the agro-food sector after
the codification
Object
s A1 A2 A3 A4 A5 D
1 1 1 1 2 3 1
2 1 2 1 2 3 1
3 1 3 1 2 3 2
4 1 4 1 2 3 1
5 1 3 1 2 3 1
6 1 4 2 2 3 1
7 1 4 3 2 3 1
8 1 1 2 2 2 1
9 1 4 1 2 2 1
10 1 4 2 2 2 1
11 1 5 3 3 2 1
12 2 1 4 1 1 2
13 2 1 4 2 1 2
14 2 6 4 3 1 1
15 2 6 4 1 1 2
16 2 6 4 2 1 1
17 2 6 4 3 1 2
18 2 1 3 3 1 4
19 2 2 3 3 1 4
20 2 3 3 3 1 4
21 1 1 1 1 3 1
22 1 1 1 1 2 1
23 1 1 1 3 2 1
24 1 5 5 3 2 1
25 1 1 1 1 3 1
26 1 1 1 1 3 1
27 1 1 1 3 1 1
28 2 1 6 2 3 3
29 1 1 6 2 3 2
30 2 1 6 2 3 3
31 2 1 6 2 3 4
32 1 1 6 2 3 3
33 3 1 6 2 3 3
34 2 1 7 3 2 4
35 2 1 7 1 2 4
36 1 1 7 4 1 2
37 4 1 7 4 1 2
38 4 1 7 4 1 2
39 1 1 7 4 1 2
40 4 1 7 4 1 2
41 4 1 7 4 1 2
42 2 1 7 4 1 2
43 2 1 7 4 1 2
44 4 1 7 4 1 2
45 2 1 7 4 1 2
46 3 1 7 4 1 2
47 4 1 7 4 1 2
48 4 1 7 4 1 1
49 4 1 7 4 1 2
50 2 1 7 4 1 2
Table 7 Accuracy and quality of classification of the decision variable for the
farm sector
Class of decision
Class of Lower Upper
decision Accuracy of approximation approximation
dependant approximati number of
variable on number of objects objects
1 0.500 7 14
2 1.000 31 31
3 0.611 11 18
Accuracy of approximation : 0.777
Quality of approximation : 0.875
Core of attributes : A1, A2, A3, A4, A5
Table 8 Accuracy and quality of classification of the decision variable for agro-
food sector
Class of decision
Lower Upper
Class of decision Accuracy of approximation approximation
dependant approximati number of
variable on number of objects objects
1 0.5862 17 29
2 0.4167 10 24
3 0.1667 1 6
4 0.6250 5 8
Accuracy of classification : 0.4925
Quality of classification : 0.6600
Core of attributes : A1, A2, A3, A4
Results:
To utilise these results there is a need to develop a classification algorithm, to
then allow for predictions to be made. The algorithm may also be used for the
classification of new values. It was noted that not all of the rules are equally
important or reliable. In the case study there were 23 exact rules and 3
approximate rules for the Total Food Productivity and 13 exact rules and 5
approximate rules for the agro-food sector. As many of the rules are based only
on observations, the quality can be quite low. To decide which rules were most
relevant, they were compared using a strength method to simplify the decision
table to a few rules.
Conculsion:
This paper shows the potential for rough set theory to allow for future predictions
of growth in the farm sector. This could be applied to allow for predictions of
yields of crops for the year to provide a better insight into the amount of food
being produced, and thus the amount needed to import/ available to export.