Profitability assessment of precision agriculture applications – a step forward in farm management

Abstract

Profitability is an underestimated concept in precision agriculture. In this research, a new module is developed within a pre-existing farm management system to assess the profitability of precision agriculture applications in extended crops. The module is regulated on a 5-meter spatial resolution, thus allowing scaling up of original and processed data on a zone-, field-, cultivar-, and farm-scale. A bottom-up approach, taking advantage of the full functionality of the farm management system, together with a flexible architecture and an easy-to-use interface, renders the new module an innovative commercial application.

Full text

Technical Note Not peer-reviewed version
Profitability assessment of
precision agriculture applications
–
a step forward in farm
management
Christos Karydas * , Myrto Chatziantoniou , Ourania Tremma , Alexandros Milios , Kostas Stamkopoulos ,
Vangelis Vassiliadis , Spiros Mourelatos
Posted Date: 5 July 2023
doi: 10.20944/preprints202307.0289.v1
Keywords: precision agriculture; site-specific fertilization; ifarma; PreFer
Preprints.org is a free multidiscipline platform providing preprint service that
is dedicated to making early versions of research outputs permanently
available and citable. Preprints posted at Preprints.org appear in Web of
Science, Crossref, Google Scholar, Scilit, Europe PMC.
Copyright: This is an open access article distributed under the Creative Commons
Attribution License which permits unrestricted use, distribution, and reproduction in any
medium, provided the original work is properly cited.

Article
Profitability assessment of precision agriculture
applications – a step forward in farm management
Christos Karydas 1,*, Myrto Chatziantoniou 2, Ourania Tremma 3, Alexandros Milios 4, Kostas
Stamkopoulos 2, Vangelis Vassiliadis 2 and Spiros Mourelatos 1
1 Ecodevelopment S.A., Filyro P.O. Box 2420, 57010 Thessaloniki, Greece
2 Agrostis S.A., VEPE Technopolis - Building C2, 55535, Pylea, Thessaloniki, Greece
3 University College Dublin, School of Agriculture and Food Science, Belfield, D04 V1W8, Dublin, Ireland
4 New Agriculture New Generation
* Correspondence: xkarydas@gmail.com; Tel: +30 2310678900 (int. 21)
Abstract. Profitability is an underestimated concept in precision agriculture. In this research, a new module is
developed within a pre-existing farm management system to assess the profitability of precision agriculture
applications in extended crops. The module is regulated on a 5-meter spatial resolution, thus allowing scaling
up of original and processed data on a zone-, field-, cultivar-, and farm-scale. A bottom-up approach, taking
advantage of the full functionality of the farm management system, together with a flexible architecture and
an easy-to-use interface, renders the new module an innovative commercial application.
Keywords: precision agriculture; site-specific fertilization; ifarma; PreFer
1. Introduction
1.1. The problem
The financial success of a business can be evaluated by its profit and profitability. Profit refers
to the absolute measure of earnings minus the expenses involved in achieving a particular outcome.
In a market-driven economic system, it is imperative for entrepreneurs to prioritize profit realization,
as failure to do so compromises the sustainability and longevity of their enterprise [1]. Apart from
maximizing profits, though, the goal of any agricultural enterprise is also to minimize costs.
Profitability represents a relative measure of a company's effectiveness, allowing for a
comparison between the achieved outcome and the associated costs [2]. To ensure the profitability of
an agricultural enterprise, efficient management is essential, typified by tasks such as soil tillage, crop
planting, irrigation, weed management, pest, and disease control, and harvesting.
Further, effective farm management requires a combination of knowledge, skills, and
experience, and often involves the use of technology and data-driven decision-making. The key lies
in effectively utilizing production resources and adopting advanced techniques to produce crops [3].
It is crucial for businesses to prioritize maximizing their profits within the limitations of available
resources, including financial and credit resources, material support for production, and the
necessary skills to carry out the tasks of the workforce [4].
Precision Agriculture (PA) is one of several methodologies which can improve farm
management by providing timely, thorough, site-specific crop information within a decision-making
framework. Data can be retrieved from a variety of sources, such as soil sampling, sensors, weather
stations, satellite or drone images, and yield monitors.
As a tool facilitating farmers to more efficiently manage their land, precision agriculture
significantly and variously impacts farm management. According to global trends, the application of
precision agriculture worldwide is estimated to increase in the next four years, with the market value
doubling from 17.41 billion U.S.＄in 2022 to 34.1billion U.S.＄in 2026 [5].
The main goal of precision agriculture research is to define a decision support system (DSS) for
whole farm management with the aim of optimizing returns on inputs while preserving resources
Disclaimer/Publisher’s Note: The statements, opinions, and data contained in all publications are solely those of the individual author(s) and
contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting
from any ideas, methods, instructions, or products referred to in the content.
Preprints (www.preprints.org) | NOT PEER-REVIEWED | Posted: 5 July 2023 doi:10.20944/preprints202307.0289.v1
© 2023 by the author(s). Distributed under a Creative Commons CC BY license.

2
[6–8]. However, farming is a complex endeavor involving many factors and inputs, such as land cost,
labor, expensive machines and various tools, fertilizers, pesticides, and irrigation. In most cases,
farming activities are not properly logged, at least not in a systematic and analytic way, while most
data are fragmented, dispersed, and difficult to use [9].
According to a recent review based on 23-year meta-analysis [10],
profitability, consultancy, and computer use had only a moderate effect on the adoption of precision
agriculture. However, these findings should be viewed cautiously due to issues of sample size and
heterogeneity embedded in some of the reference studies, while at the same time, other factors had a
negligible effect on adoption. Precision agriculture must be discriminated from “smart agriculture”,
regarding the concept of site-specificity. Smart agriculture is associated with various types of sensors
used (soil, moisture, climatic, etc.) to derive crop information (and potentially return a decision),
regardless satisfaction of within-field spatial variability, as it happens with precision agriculture
practices.
1.2. State of the art
Lately, several farm management platforms and technologies have become available to support
precision and smart agriculture applications; below are some indicative platforms in English
available in the market:
• Climate FieldView: An integrated digital platform that collects and analyzes field data, helping
farmers make more informed decisions regarding crop management, planting, and harvesting.
The available tools allow farmers to manually delineate management zones
(https://www.fieldview.com.au/).
• Granular’s Farm Management Software (FMS), credited as the first cloud-based, mobile-centric
program of its kind, offers an intuitive breakdown of everything a farmer needs to consider,
from financial to soil management to operations. The platform is mostly oriented to sensors and
smart agriculture (https://www.corteva.com/resources/media-center/granular-provides-new-
digital-nitrogen-management-options-to-farmers.html).
• Farmers Edge: A comprehensive smart agriculture platform that includes field-centric data
collection, satellite imagery, variable rate technology, and weather analytics to optimize farm
operations (https://farmersedge.ca/).
• Agworld: A collaborative farm management platform that allows farmers, agronomists, and
other stakeholders to work together on planning, budgeting, and reporting of farm activities. It
incorporates add-in applications for specific works (e.g. Satamap for satellite image display)
(https://www.agworld.com/us/).
• Taranis: This platform uses artificial intelligence (AI)-driven image analysis, combining high-
resolution aerial imagery and field-level weather data to detect and predict pest and disease
issues, enabling farmers to make proactive decisions (https://www.taranis.com/).
• Trimble: A platform offering a range of precision agriculture solutions, mostly oriented to
equipment and automation, including guidance and steering systems, flow and application
control, yield monitoring, and water management tools
(https://agriculture.trimble.com/en/products/software/trimble-agriculture-software).
• Sentera: A platform that integrates drone and satellite imagery with sensor data, enabling
farmers to monitor plant health, track growth, and identify potential issues
(https://sentera.com/).
• John Deere Operations Center: A web-based platform that helps farmers track equipment,
manage field data, and analyze agronomic information to optimize their operations
(https://operationscenter.deere.com/).
• Topcon Agriculture: A suite of visualization and decision-making tools including auto-steering
systems, variable rate control, yield monitoring, and farm management software
(https://tap.topconagriculture.com/).
• Raven Industries: Providing automations like guidance and steering systems, application
controls, and field computers to help farmers optimize their operations (https://ravenind.com/).
Preprints (www.preprints.org) | NOT PEER-REVIEWED | Posted: 5 July 2023 doi:10.20944/preprints202307.0289.v1

3
Most of the above solutions -although quite technologically advanced- respond partially or even
inadequately to the need for an integrated site-specific management. Moreover, economic records
stored in the database (if they exist) are not always effectively linked to relevant precision farming
data, or (even if they are), the platform leaves the user alone to carry out the analysis or make out the
decisions. Therefore, in all cases there is a gap between available data, decision making on site-
specific applications, and their economic evaluation.
To bridge this gap between farm management and precision agriculture applications, two Greek
enterprises, Agrostis and Ecodevelopment, cooperated in 2022 to incorporate a site-specific
fertilization service (namely PreFer) into a pre-existing Farm Management Information System
(FMIS) (namely ifarma) [9].
ifarma was introduced to the Greek market in 2014 by Agrostis, as a cloud-based farm
management information system (FMIS). The data model of ifarma integrates all information relevant
to a farm, such as fields and land parcels, crops, farming activities on fields and inputs and resources
used to plan and execute these activities. This data model organizes the information in a hierarchical
manner, where the farm is at the top. Today, ifarma is a well-known trademark recognized as the
best farm management software for agricultural holdings in Greece [11].
PreFer service developed by Ecodevelopment, produces prescription maps, which together with
a variety of spatial layers (including soil properties, agronomic information, crop indices, statistical
and predictive climatic parameters, and yield records) become available to the farmers on 5-meter
spatial resolution on a regular basis. The prescription maps are created within a GIS, where big data
are analyzed using machine learning methodologies [12].
1.3. Objectives
Going a step further, the objective of this research was to offer a complete and easy-to-use
commercial solution for profitability assessment of precision agriculture applications in extended
crops on an annual basis.
Accordingly, a new module was developed within the ifarma farm management platform using
PreFer functionalities, especially its mapping environment and algorithms, thus facilitating
interoperability, swiftness, and ease. In this respect, the next Section will present the materials and
methods employed, Section 3 will demonstrate the results along with discussion and Section 4 will
provide the main conclusions of this work.
2. Materials and Methods
2.1. System architecture
The new profitability module, namely, ProFit, is an independent module of the ifarma FMIS, in
terms of interface and algorithms, although it cooperates with the PreFer module of ifarma, for
exchanging map data. More specifically, ProFit takes spatial data from the PreFer database as input
into its algorithms and returns output maps for display in the map viewer of PreFer. In this way, the
original PreFer (say v.1) is upgraded into a new version (say v.2) after integrating with ProFit (Figure
1).
Preprints (www.preprints.org) | NOT PEER-REVIEWED | Posted: 5 July 2023 doi:10.20944/preprints202307.0289.v1

4
Figure 1. The architecture of ProFit, based on its conjunction with PreFer module of ifarma.
Going a step further, the objective of this research was to offer a complete and easy-to-use
commercial solution for profitability assessment of precision agriculture applications in extended
crops on an annual basis.
ProFit alone comprises two components (Figure 2). A site-specific cost component, which is fed
by the database of PreFer, where the precision agriculture applications are stored and displayed. A
shared cost component for other (non-precision agriculture) practices, where the required records are
manually entered as lump sum amounts. The site-specific cost component takes input from two kinds
of spatial data: a) the fertilizer application maps and b) the yield maps. The former is used to calculate
fertilization cost at every surface unit (of 25 m2) after the multiplication of the fertilizer’s rates with
its corresponding unit cost; the latter, meanwhile, are used to calculate earnings at every surface unit
after the multiplication of the yield with the price of the corresponding cultivar in the market.
The shared cost component holds the lump sum amounts per expenditure category. The
distribution of these lump sum costs is then based on an empirical classification of the fields of the
farm according to a degree of difficulty (or weighting factor) on a categorical scale of 1-5, with ‘1’
corresponding to the easiest field and ‘5’ to the most difficult for each of the agriculture practices
applied (e.g., soil tillage, irrigation, weed management, etc.).
The output data will be in two forms: a) descriptive statistics of cost, earnings, and profit per
field; and b) profitability maps (cost, earnings, and profit maps) at a 25-m2 surface unit. The
calculations will be done automatically according to embedded formulas.
Figure 2. The structure of the profitability module in ifarma (partially functioning within the PreFer
module).
Preprints (www.preprints.org) | NOT PEER-REVIEWED | Posted: 5 July 2023 doi:10.20944/preprints202307.0289.v1

5
2.2. Data requirements
The data entry of the shared cost records in ProFit is carried out manually through a new form
which is divided into two sections, each corresponding to a number of agricultural practices. The site-
specific data will be read from the maps stored in the PreFer database.
The shared cost data are related to total annual amounts for the entire cultivation and can be
divided into the following categories (using a common ordering regardless change of category):
Section A (related to categorized total annual costs):
1. Land rent
2. Seeds
3. Irrigation
4. Fertilizers
5. Weed killers
6. Pesticides/Insecticides
7. Harvest
8. Machinery
a) Depreciation
b) Maintenance
c) Spare parts
Section B (related to total annual costs per field):
9. Land rent (absolute amounts)
10. Degree of difficulty per field for shared cost (weighting factor: 1-5)
d) Seeds
e) Irrigation
f) Fertilizers
g) Weed killers
h) Pesticides/Insecticides
i) Harvest
j) Machinery
2.3. Algorithms developed
A set of interrelated functions have been setup for ProFit, which carry out arithmetic, categorical,
and logical operations, and transfers of tabular data. The overall arrangement is integrated and
executed in a prototype Excel spreadsheet, where all internal functions and options and the external
data feed are arranged. The monetary rate unit is set to euros per hectare (€/ha), the fertilizer amounts
in kilograms per hectare (kg/ha), and the yield in tonnes per hectare (t/ha) (Figure 3).
Preprints (www.preprints.org) | NOT PEER-REVIEWED | Posted: 5 July 2023 doi:10.20944/preprints202307.0289.v1

6
Figure 3. The prototype spreadsheet of ProFit algorithm, filled with true data from the 2022 cultivation year (4 indicative fields).
[2]
[3]
[4]
[5]
[6]
[7]
[8]
[9]
[10]
[11]
ANNUAL TOTALS (€)460 3,392 9,162 2,925 542 10,033 2,950 740 3,312 1,670
FIELDS CULTIVARS SURFACE (ha) [1]
LAND RENT (€/ha)
RATE OF DIFFICULTY FERTILIZERS UNIT COST
(€/kg)
COST
(€/kg)
006
CL 111
3.79
950
1
1
2
5
3
1
2
1
1
1
ALZON
1.00
006
ALZON
006
007
CL 111
3.75
1,010
1
1
2
1
2
1
2
1
2
1
TSP
0.73
006
TSP
007
008
RONALDO
4.37
1,100
2
1
1
3
1
1
1
1
2
2
K2SHO4
0.85
006
K2SHO4
008
009
RONALDO
5.05
1,050
2
1
1
2
1
1
1
1
4
1
N40
0.78
006
N40
009
27
200
540
172
32
592
174
44
195
98
007
ALZON
007
TSP
007
K2SHO4
AMOUNTS (€)
007
N40
FIELDS CULTIVARS SURFACE (ha) [1] RENT COST (€) [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] TOTAL COST
(€)
UNIT COST
(€/ha)
EARNINGS
(€)
PROFIT €)008 ALZON
006
CL 111
3.79
3,601
12,613
3,328
20,193
7,580
008
TSP
007
CL 111
3.75
3,788
65
239
144
2,218
903
164
617
294
11,986
3,196
19,800
7,814
008
K2SHO4
008
RONALDO
4.37
4,807
152
874
835
84
2,585
526
191
720
684
13,092
2,996
27,610
14,517
008
N40
009
RONALDO
5.05
5,303
176
1,010
643
97
2,987
608
220
1,663
395
14,991
2,969
32,724
17,733
009
ALZON
16.96
009
TSP
4.24
47.5%
009
K2SHO4
PROTIT RATIO
009
N40
FIELDS CULTIVARS SURFACE (ha) P RICE (€/kg) MEAN YIELD PER CULTIVAR
(t/ha)
INCOME (€) COST (€) PROFIT (€)UNIT PROFIT
(€/ha)
006 & 007
CL 111
0.48
11.1
39,993
24,599
15,394
2,042
008 & 009
RONALDO
0.54
11.9
60,334
28,084
32,250
3,424
52,683
2,809
INPUT
SHARED COST
BALANCE SHEET PER FIELD
BALANCE SHEET PER CULTIVAR
100,327
47,644
CATEGORIES
SEEDS
IRRIGATION
FERTILIZERS
WEED KILLERS
PESTICIDES &
INSECTICIDES
HARVEST
ΕQUIPEMENT
DEPRECIATION
EQUIPEMENT
MAINTENANCE
FUELS
LABOUR
TOTAL COST (€)
MEAN APPLIED
UNIT (kg/ha)
TOTAL APPLIED
AMOUNT (kg)
MEAN
YIELD
(t/ha)
TOTAL
YIELD (t)
FIELD
FERTILIZERS
FIELD
CULTIVAR
6,610
390
1,478
CL 111
11.1
42.1
855
81
307
CL 111
11.0
41.3
1,697
150
569
RONALDO
11.7
51.1
3,043
230
872
RONALDO
12.0
60.6
TOTAL SURFACE
AVERAGE UNIT COST (€)
TOTAL
9,162
390
1,463
OVERALL
11.5
195.0
NUMBER OF FIELDS
4
TRUE
TRUE
TRUE
TRUE
TRUE
TRUE
TRUE
TRUE
TRUE
TRUE
TRUE
AVERAGE
540
59
221
127
476
230
863
389
1,700
66
758
2,835
1,207
218
2,242
913
165
312
297
58
253
750
2,805
100
437
1,634
230
1,005
1,888
390
1,970
TOTALS
17,498
460
3,392
9,162
2,925
542
10,033
2,950
740
3,312
1,670
77
389
AVERAGE
1,032
27.1
200.0
540.2
172.5
32.0
591.5
173.9
43.6
195.3
0.1
FALSE
3,106
TRUE
102
515
230
1,162
OVERALL
806
13,678
TOTAL
53,749
FERTILIZERS
TOTAL APPLIED
AMOUNT (kg)
MEAN APPLIED
UNIT (kg/ha)
7.54
/ha
3,169
ALZON
6,610
390
9.42
TSP
1,171
69
TOTALS
K2SHO4
1,997
118
N40
3,901
230
OUTPUT
TOTAL
13,678
TRUE
SITE-SPECIFIC DATA INPUT (MAPS)
51,013
SITE-SPECIFIC COST
FERTILIZATION MAPS
YIELD MAPS
16.96
52,683
100,327
47,644
SUBSIDIES
€/ha
€
GREENING
300
5,088
RIGHTS
60
1,018
TOTAL
6,106
Preprints (www.preprints.org) | NOT PEER-REVIEWED | Posted: 5 July 2023 doi:10.20944/preprints202307.0289.v1

7
The basic formula for cost, earnings, and profit calculations applied per field, is shown simplified
below:
Cost =
Field Rental Cost +
Field Shared Cost for Agricultural Practice [1] +
Field Shared Cost for Agricultural Practice [2] +
Field Shared Cost for Agricultural Practice [3] +
…
Field Shared Cost for Agricultural Practice [8c] +
Quantity of Fertilizer-1 [Input from the Fertilization Maps of PreFer database] x Price of Fertilizer-1
(manual input in relevant table) +
… (as many times as the number of fertilizers, n)
Quantity of Fertilizer-n [Input from the Fertilization Maps of PreFer database] x Price of Fertilizer-1
(manual input in relevant table)
Earnings = Yield (t/ha) [Input From the Yield Map] x Cultivar Price (monetary unit/kg) [Input from
the Crop List]
Profit = Earnings – Cost
For splitting the shared cost lump sums of the different agricultural work categories into cost
items per field, a stepwise procedure is followed (Figure 4):
1. First, the rate of difficulty of each field is multiplied by the field’s extent and then divided
by the number of fields under consideration, to give a weighted rate of difficulty.
2. Then, the weighted rate of difficulty of each field is divided by the total weighted rate of
difficulty to give the cost share for the field (for each of the shared cost categories).
3. Finally, the cost share of every field is multiplied by the total cost of the category and
divided by the number of fields to give the absolute cost per field (for that cost category).
Therefore, for every category, the algorithm retrieves the lump sum amount for the entire
cultivation from a manually filled table and the field extents from the farm’s geodatabase (i.e. from
PreFer). Thus, the only inputs required by the algorithm are the rates of difficulty per field and for
each work category. An internal control function will check if the earlier entered lump sum for the
working cost category is equal to the one calculated by the algorithm.
Figure 4. An example of the cost lump sum splitting algorithm (prototype) (same dataset as Figure 3).
3. Results and Discussion
3.1. System functionality
The new module ProFit comprises a spatial and a non-spatial component. The spatial component
is developed in the pre-existing module PreFer within the ifarma FMIS. As a result, the output maps
of ProFit follow the same standards as that of the PreFer module, i.e., a spatial resolution of 5 meters
(surface unit of 25 m2) and a classification of the original values into 7 categories.
The non-spatial components include output tables which at the same time contain the entire
input information. Thus, the user gets the whole economic picture of the cultivation in a single tabular
arrangement.
Total cost of work (€)
Rate of difficulty (1-5)
Weighted rate
Cost share
Cost per field (€)
Unit field cost (€)
Relative cost
4.7
1.65
1207
318.6
184.7%
0.9
0.33
239
63.7
36.9%
3.3
1.14
835
191.1
110.8%
2.5
0.88
643
Totals
16.96
2.9
1.00
Number of fiel ds
4
Control
TRUE
Average unit cost (€)
172.5
2,925
FIELDS
SURFACE (ha)
006
3.79
5
007
3.75
1
008
4.37
3
009
5.05
2
2,925
Preprints (www.preprints.org) | NOT PEER-REVIEWED | Posted: 5 July 2023 doi:10.20944/preprints202307.0289.v1

8
3.2. System interface
The ProFit module is available as an option in the main menu of the ifarma interface, specifically
in the ‘Management of works, inputs, and yields’ group (currently only in Greek). By clicking the
ProFit button (set below the ‘PreFer’ option), the input form for entering values for the required
economic items is launched in a single web page (Figure 5). For convenience, the output fields are
displayed in a different form after the execution of the calculations and can be exported to
spreadsheets.
Apart from the tabular output data (e.g., statistics per field and the entire cultivation), which are
displayed on a different web page of the ifarma environment, the output maps are displayed within
the PreFer map viewer. The options for displaying the cost, earnings, and profit maps for every
cultivation are available inside the ‘Performance’ group of options in the PreFer map viewer menu
(which also includes the yield maps).
3.3. Experiences
Most of the existing platforms in the market are top-down, originating from the need of an
already established market player to promote its products, services, or goods through the concepts
of precision agriculture. In most cases, these platforms provide financial management as a parallel
service to another main service, e.g., fleet management, crop calendar, etc.
Conversely, ProFit on ifarma follows a bottom-up approach, starting from the applications and
considering all the farming details to conduct a thorough economic analysis, while enabling spatially
distributed outputs in terms of maps.
The development team was assisted significantly by several farmers who implemented precision
agriculture over the course of years, through their ideas and experience and by using and testing the
module with authentic farming data from the 2022 cultivation year in Greece.
Using real data combined with some notional options, for example, in the selection of the
difficulty factors per field, to test the module under extreme data ranges it was noticed that yield
maps might be quite different from profitability maps (especially between fields) (Figure 6).
Preprints (www.preprints.org) | NOT PEER-REVIEWED | Posted: 5 July 2023 doi:10.20944/preprints202307.0289.v1

9
Figure 5. The input form for the required data for profitability assessment by ProFit (ifarma interface).
Preprints (www.preprints.org) | NOT PEER-REVIEWED | Posted: 5 July 2023 doi:10.20944/preprints202307.0289.v1

10
a
b
Figure 6. An example of yield vs. profit map of the same fields and year in the ifarma/ProFit
environment (example of Mr. Kostas Kravvas rice farm, Greece).
5. Conclusions
In this research, an innovative module for assessing the profitability of precision agriculture
applications was developed, namely ProFit. In terms of architecture, ProFit is embedded within a
cloud-based farm management information system, namely “ifarma”, while taking advantage of pre-
existing functionalities, such as a precision agriculture database provided by other ifarma modules,
like PreFer.
ProFit offers an easy-to-use interface, which encourages farmers to enter economic records
quickly and reliably, while using an empirical method to share apportionable expenditures between
fields (when and where site-specific maps are not available). At the same time, it uses the map view
environment of PreFer to display the profitability maps.
Future work will focus on widening the range of precision agriculture practices (I.e., beyond
fertilization) within ProFit, such as soil tillage, seeding, irrigation, weed management, and crop
protection.
Author Contributions: Conceptualization, C.K., M.C., and S.M.; methodology, C.K., M.C., and V.V.; software,
K.S. and V.V.; validation, M.C., A.M., and K.S.; formal analysis, C.K. and M.C.; investigation, M.C. and O.T.;
resources, M.C. and O.T.; data curation, C.K. , A.M. and K.S..; writing—original draft preparation, C.K., M.C.,
and O.T.; writing—review and editing, M.C., and O.T.; visualization, A.M., and K.S; supervision, S.M. and V.V.;
project administration, V.V.; funding acquisition, S.M. All authors have read and agreed to the published version
of the manuscript.
Funding: This research received no external funding.
Informed Consent Statement: Not applicable.
Data Availability Statement: Data will be made available on request.
Acknowledgments: Special thanks to Mr. Kostas Kravvas and Mr. Panagiotis Goutas, rice farmers in Greece, for
their ideas and the permission to use their data for the demonstration needs of this work.
Conicts of Interest: The authors declare no conict of interest.
References
1. Barkley, A.; Barkley, P. W. Principles of Agricultural Economics, 1st ed.; Routledge: Abingdon-London,
United Kingdom, 2013, pp. 1-384. DOI: https://doi.org/10.4324/9780203371145
2. Epifanova, I. Y.; Humeniuk, V. S. Profitability of the enterprise: modern approaches to defining the essence.
Econ Soc 2016, 3, 189-192.
Preprints (www.preprints.org) | NOT PEER-REVIEWED | Posted: 5 July 2023 doi:10.20944/preprints202307.0289.v1

11
3. Smoliy, L.V. Formation of the results of economic activity of Agricultural Enterprises. Collected Works of
Uman National University of Horticulture 2019, 95(2), 115-128. DOI: 10.31395/2415-8240-2019-95-2-115-128
4. Shmatkovska, T. О.; Dziamulych, M.; Vavdiiuk, N.; Petrukha, S.; Koretska, N.; Bilochenko, A. Trends and
Conditions for the Formation of Profitability of Agricultural Enterprises: A Case Study of Lviv Region,
Ukraine. Univers. J. Agric. Res. 2022, 10(1), 88–98. DOI: https://doi.org/10.13189/ujar.2022.100108
5. Shahbandeh, M. Smart Agriculture: Market Size Worldwide 2026.
https://www.statista.com/statistics/720062/market-value-smart-agriculture-worldwide/ (accessed on
20/04/2023).
6. McBratney, A. B.; Whelan, B.; Ancev, T.; Bouma, J. Future Directions of Precision Agriculture. Precis. Agric.
2005, 6(1), 7–23. DOI: https://doi.org/10.1007/s11119-005-0681-8
7. Whelan, B.; McBratney, A. B. Definition and interpretation of potential management zones in Australia.
Solutions for a Better Environment. In Proceedings of the 11th Australian Agronomy Conference, Geelong,
Victoria, Australia, 2-6 February 2003, 0–11.
8. Singh, P.; Pandey, P. C.; Petropoulos, G. P.; Pavlides, A.; Srivastava, P. K.; Koutsias, N.; Deng, K. A. K.; Bao,
Y. Hyperspectral remote sensing in precision agriculture: present status, challenges, and future trends. In
Earth Observation, Hyperspectral Remote Sensing, Pandey, P. C.; Srivastava, P. K.; Balzter, H.; Bhattacharya,
B.; Petropoulos, G. P.; Elsevier eBooks, 2020, pp. 121–146. https://doi.org/10.1016/B978-0-08-102894-0.00009-
7
9. Karydas, C. G.; Chatziantoniou, M.; Stamkopoulos, K.; Iatrou, M.; Vassiliadis, V.; Mourelatos, S.
Embedding a precision agriculture service into a farm management information system- ifarma/PreFer.
Smart Agric. Technol. 2023, 4, 100175. DOI: https://doi.org/10.1016/j.atech.2023.100175
10. Tey, Y.S.; Brindal, M. A meta-analysis of factors driving the adoption of precision agriculture. Precis. Agric.
2022, 23, 353–372. DOI: https://doi.org/10.1007/s11119-021-09840-9
11. Paraforos, D.S.; Vassiliadis, V.; Kortenbruck, D.; Stamkopoulos, K.; Ziogas, V.; Sapounas, A. A.;
Griepentrog, H.W. Multi-level automation of farm management information systems. Comput. Electron.
Agric. 2017, 142, 504–514, DOI: https://doi.org/10.1016/j.compag.2017.11.022
12. Iatrou, M.; Karydas, C.; Tseni, X.; Mourelatos, S. Representation Learning with a Variational Autoencoder
for Predicting Nitrogen Requirement in Rice. Remote Sens. 2022, 14, 5978. DOI:
https://doi.org/10.3390/rs14235978
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those
of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s)
disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or
products referred to in the content.
Preprints (www.preprints.org) | NOT PEER-REVIEWED | Posted: 5 July 2023 doi:10.20944/preprints202307.0289.v1