International Journal of Sustainable Agricultural Management and Informatics (14 papers in press)

Regular Issues

• Factors impacting farm management decision making software adoption  
  by Kenneth David Strang, Sarah Bitrus, Rao Vajjhala 
  Abstract: In this study, we use an unconventional socio-cultural ideology to examine if Western African farmers think agricultural information systems (AISs) improve economic production, at the individual level of analysis. Food production is a necessity for our survival but it has been negatively impacted by unstable financial markets, climate change, political upheavals and health pandemics, especially in Western African-based developing countries. We developed a four factor model based on the information systems expectancy confirmation theory which could determine why Nigerian farmers adopt agricultural information systems. We used a survey to collect data from farmers, we validated the questions using a pilot study, and we developed a structural equation model to quantitatively explain why farmers make the decision to adopt or discontinue the use of AIS electronic software. We found that farmers satisfaction was positively influenced by high confirmation experience with AIS. To a lesser extent we found that farmers satisfaction was positively impacted by AIS perceived usefulness (PU). Interestingly, we found no evidence that any factor was related to farmers behavioural intent for continued use of AIS. The results raise controversial issues concerning AIS effectiveness and government technology funding in Western African countries.
  Keywords: agricultural information system; AIS; farm management; crop planning; decision making software; adoption; perceived use; continuance intention; satisfaction; confirmation experience; expectancy confirmation theory; Western Africa; Nigeria.
  DOI: 10.1504/IJSAMI.2019.10019819
   
  
• Selection of the most suitable tree species in urban areas based on their capability of capturing heavy metals. A Forest Policy approach.  
  by Stefanos Tsiaras, Theano Samara 
  Abstract: The aim of this paper to select the most suitable tree species in urban areas, taking into consideration their capability of capturing heavy metals. The selection is based on a broadly accepted tool for Forest Policy, Multiple Criteria Decision Analysis and more specifically, the PROMETHEE method. Five of the most common tree species for urban green in Greece (Cupressus arizonica, Albizia julibrissin, Celtis australis, Platanus orientals and Ligustrum japonicum) were examined, and the criteria were the captivity of seven heavy metals: manganese (Mn), zinc (Zn), copper (Cu), lead (Pd), cadmium (Cd), chromium and nickel (Ni). All heavy metals were considered of equal importance and therefore, the weights of the criteria were equal. According to the PROMETHEE ranking the best choice among the alternatives is the Arizona cypress; this is a reasonable outcome, as the cypress is an evergreen species and it absorbs heavy metals during the whole year, in contradiction with the deciduous tree species. Broadleaves species with compound leaves, such as Albizia julibrissin could also capture significant amount of heavy metals especially the manganese. The selection of these species would have important benefits for the citizens of Thessaloniki, a city characterized by its densification and lack of urban green. Forest Policy can play an important role in planning the green spaces in cities in an attempt to increase urban green, a vital decision with many benefits for the citizens.
  Keywords: Air pollution; urban green; Multiple Criteria Decision Analysis; PROMETHEE; Forest Policy.
  
  
• Soil Moisture Sensors for Sustainable Irrigation: Comparison and Calibration  
  by Vaibhav Bhatnagar, Ramesh C. Poonia, Jagdish Prasad 
  Abstract: Soil moisture is a vital component for plant growth. There are many electro-chemical low-cost sensors that sense soil moisture with immediate effect rather than other methods such as Gravimetric Method and others that are expensive as well as consume lot of time. Some of these electro-chemical sensors work on resistivity whereas some sensors work on capacitance. This paper compares the efficiency of resistivity based sensors with capacitance based soil moisture sensors so that farmers and experts can use the better one for designing IOT based irrigation monitoring and controller system. This comparison is performed with twenty one samples having different levels of soil water content. The Coefficient of Variation is calculated to compare the efficiency of both the sensors. After the comparison, calibration has also been performed that depicts the soil moisture content to LCD Display and alert the farmer by a Buzzer.
  Keywords: IOT; Soil Moisture Sensor; Arduino; Calibration.
  
  
• Satellite methodologies for rationalizing crop water requirements in vulnerable agroecosystems  
  by Nicolas Dalezios, Anna Blanta, Athanasios Loukas, Marios Spiliotopoulos, Ioannis Faraslis, Nicholas Dercas 
  Abstract: Vulnerability in agriculture is influenced, among others, by extended periods of water shortage in regions exposed to droughts. In order to assess irrigation water requirements, remote sensing techniques are integrated for the estimation and monitoring of cotton crop evapotranspiration ETc. Cotton fields in a small agricultural sub-catchment in Thessaly, central Greece, are used as an experimental site. Daily meteorological data and weekly field data are recorded throughout seven (2004-2010) growing seasons for the computation of reference evapotranspiration ETo, crop coefficient Kc and cotton crop ETc based on conventional data, which are compared during the corresponding period with the satellite-based method (Landsat TM) for the estimation of cotton crop coefficient Kc and cotton crop ETc, which also delineates its spatiotemporal variability. The methodology is applied for monitoring Kc and ETc during growing season in the selected sub-catchment. Several error statistics are used showing very good agreement with ground-truth observations.
  Keywords: Crop evapotranspiration; remote sensing; crop water needs; vulnerable agriculture.
  
  
• Analysis and Optimization of an Olive Oil Supply Chain: A Case from Turkey  
  by Oznur Yurt, Laszlo Kato, Karoly Jarmai, Ebru Aglamaz 
  Abstract: The aim of this study is to determine potential improvement areas for food supply chain operations. For this purpose, a model is developed to optimize the distribution network of an olives and olive oil company in Turkey. The current distribution system of the company is analyzed and a mathematical programming model is developed to provide a distribution design to maximize the profit. An integer programming model is used and the problem is solved by using CPLEX solver/GAMS Software. Production volumes for each product type proposed by the results are different than the real production volumes. Based on the optimal results proposed by the solution, required changes in the production volumes are given. Similarly, distribution of optimal quantities of each product to each province is different than the real data. Results also suggest that, production and distribution network decisions for four different product groups must be reconsidered and production and distribution systems of the company have to be redesigned. Although this paper focuses on a single case, the model proposed in this study and findings of this study provide guidance for food supply chain members with similar problems.
  Keywords: supply chain; food supply chain; olive-oil industry; distribution network optimization; Turkey.
  
  
• Current trends and challenges in the deployment of IoT technologies for climate smart facility agriculture  
  by Eleni Symeonaki, Konstantinos Arvanitis, Dimitrios Piromalis 
  Abstract: Climate Smart Facility Agriculture is considered to be a critical factor in terms of sustainability due to the predictions of the world population increase. Since the cutting edge technology of the Internet of Things (IοT) was introduced as the next Internet revolution, enabling its continuously extending applications in facility agriculture is expected to become a major asset. In particular, the implementation of the IoT technologies in facility agriculture, through the intelligent monitoring and automated control of the entire agricultural production and food chain, is an innovative research field of essential importance for the global sustainable growth. In this paper an attempt is made to survey the most significant approaches regarding the technologies and applications of the IoT in the sector of facility agriculture as well as to identify the trends and challenges regarding their efficient deployment in the context of climate smart philosophy for the benefit of sustainable development.
  Keywords: Internet of Things; smart agriculture; sustainability; food safety; intelligent monitoring; intelligent control; automation; Wireless Sensor Networks.
  
  
• HYBRID APPROACH FOR FRUITS QUALITY PREDICTION USING IMAGE PROCESSING AND SENSORS TECHNIQUE  
  by Amol Bandal, Mythili Thirugnanam 
  Abstract: The ability to recognize freshness of fruits and vegetables will be an incredible help for agriculturists to optimize the harvesting stage which abstains from reaping either under-developed or over-developed natural products. Techniques like image processing, NIRS (Near Infrared Spectroscopy), sensors are used to predict the quality of fruits. Most of these mentioned techniques are costly and focused on external features of the fruits. Therefore this paper presents a hybrid framework to predict the quality of fruits by combining image processing techniques and sensor setup. The proposed hybrid framework improves prediction accuracy rate by considering external as well as internal features of the fruit. Applying image preprocessing, 36 features are extracted from all banana samples. Among them, the most 5 influential features selected with the help of correlation matrix and linear regression. Simultaneously, with the help of multi-sensor setup, 4 influential gaseous features are extracted. K-means unsupervised learning algorithm applied on effective features of banana samples and their quality classified into three categories, such as Eatable, MaybeEatable, NotEatable. The implemented framework could classify banana fruit samples with approximately 95% accuracy.
  Keywords: Banana; Fruits classifications; Image processing; Sensors; clustering; K-Means; regression analysis.
  
  

Special Issue on: AI2SD'2018 Advanced Intelligent Systems for Sustainable Development Applied to Agriculture and Environment

• Investigating the impact of drying parameters on the dandelion root using full factorial design of experiments  
  by Haytem MOUSSAOUI, Hamza Lamsyehe, Ali Idlimam, Abdelkader Lamharrar, Mounir Kouhila 
  Abstract: Solar drying is regarded as the most convenient technique of conservation in the preservation field. This paper aims at studying the impact of drying parameters on the dandelion root as well as analyzing the interactions between the factors based on factorial experiments, statistical calculations, and ANOVA analysis. The drying of the dandelion plant process was carried out in an indirect solar dryer with a separate solar collector and a drying unit. The outcome is a model that helps to analyze the impact of drying parameters on the response, which is the time of drying. Determining the factors that affect the response is a prerequisite step so that we can analyze their influence on the drying time and on the quality of the sample.
  Keywords: Solar energy; Dandelion root; drying parameters; semi-empirical models; ANOVA; full factorial design.
  
  
• Predictable consequences of climate change for varieties of strawberry plants grown in Morocco  
  by Mohammed Ezziyyani, Ahlem Hamdache, Mostafa Ezziyyani, Loubna Cherrat 
  Abstract: The cultivation of strawberry in Morocco has developed remark-ably during the last 20 years. During the 2016-2017 crop year, this crop covers 3.050 hectares of land, including 180.378.742 straw-berry plants imported from various varieties: Sabrina, San Andreas, Fortuna, Festival, Camarosa, Splendor and others. The period from 1990 to 2010, the dominant varieties that were grown are Chandler, OsoGrande and especially Camarosa and this thanks to its very high productivity, profitability, precocity, quality and adaptation to agrocli-matic conditions of the perimeter of Luokkos. Moreover, from 2010, the Californian variety Camarosa (and others) experienced a dramatic decline. Farmers had lost patience because of low yields (very low productivity <500g / plant) and doubts were starting about the choice of the variety. This upset the choice of the distribution of varieties of strawberry plants imported in 2017. Today, many varieties are disap-pearing Moroccan producers, the choice being dictated by the produc-tion objectives.
  Keywords: Climate change; Strawberry; Interannual variability; production; adaptation; Sabrina; San Andreas; Fortuna; Festival; Camarosa; Splendor; Loukkous.
  
  
• Type 2 fuzzy TOPSIS for agriculture MCDM problems  
  by Mohamed EL ALAOUI, Khalid El Yassini, Hussain Ben-azza 
  Abstract: Multi Criteria Decision Making methods achieve an unprecedented success in various domains. However, they are not infallible. TOPSIS, one of the most used, does not check the decision makers assessments consistency nor their level of confidence. To take account of the preeminent uncertainty, we propose a TOPSIS based method with interval type 2 fuzzy numbers. An improved algorithm is presented to take into account both coherence and confidence level. An illustrative example treating an agriculture supply chain demonstrates a better sorting capacity of the proposed methodology. In addition, the suggested aims preventing rank reversal.
  Keywords: TOPSIS; coherence; reliability; interval type 2 fuzzy numbers; MADM; sustainable agriculture.
  
  
• Comparative study of the pathogenic variability of some Morrocan isolates of Botrytis cinerea  
  by Ahlem Hamdache, Ahmed Lamarti 
  Abstract: Botrytis cinerea attacks economically important crops such as strawberries producing grey mould. In this work, our aim was to select the isolate the most pathogen of B. cinerea to use it in biological control tests. For this purpose, we have studied, in vitro, the growth and the sporulation of some isolates of Botrytis cinerea in different nutrient media. Then we have investigated the effect of some bacterial isolates on these botrytis populations isolated from strawberry fruits. Seven strains were isolated from samples of strawberry fruit harvested from fields; four of them, are originated from strawberry fields of Loukkos and three are originated from fields of Moulay Bouselham. Others are isolated from postharvest strawberry. Among all of the isolates tested, Botrytis cinerea Bt7 originated from fields of Zlaoula (Loukkos) was the most important isolate with maximum growth and sporulation. Bt7 was the most resistant isolate to the antifungal activity of the antagonistic bacteria.
  Keywords: Botrytis cinerea; nutrient media; mycelial growth; sporulation; pathogenicity; Moulay Bouselham; Loukkos; antifungal activity; antagonism.
  
  
• Modelling potential impacts of climate change on the geospatial distribution of phytopathogenic telluric fungi.  
  by Mohammed Ezziyyani 
  Abstract: A very likely consequence of global warming would be a change in the range of some phytopathogens such as Phytophthora capsici, Rhizoctonia solani and Fusarium oxysporum. certain microorganisms to have a distribution limited by temperature.In this study, in vitro we focused on the mean rate of mycelial growth as a function of the time (Vmax=d/t) of the three phytopathogens, at three different temperatures (20, 25 and 30
  Keywords: climate change; temperature; pathogenicity; telluric phytopathogenic fungi; virulence; Phytophthora capsici; Rhizoc-tonia solani and Fusarium oxysporum; geospatial distribution.
  
  
• how can data mining help us to predict the influence of climate change on mediterranean agriculture?  
  by MAROI TSOULI FATHI, Mostafa EZZIYYANI 
  Abstract: Agriculture is an important sector for Morocco's sustainable devel- opment in Morocco's new strategic plan for the development of agriculture and rural development. The proposed works are part of the Moroccan national project "Green Moroc- co", which aims to develop the agricultural sector to prevent famine and support the economic development of the country. the objective is to propose a new method for yield estimation, crop forecasts and pre-harvest yields. this new ag- ricultural yield forecasting methodology is developed, introducing unsupervised learning and data mining algorithms. the method used, predicting yields using agro-climatic data analysis and climate change predictions. Our approach is based on three models: first, we propose an approach based on the modern representation of the ideology of supervised learning. the goal is to use climate change-based climate predictions for climate classification in a giv- en region and analyze the changes that have affected this climate Second, using climate rules, we will look for climatic rules for each crop. thirdly, to put in place a predictive model that will later enable climate change adaptation solutions for the future. We are also introducing a new dimensionality reduction technique that allows us to network and learn more about it. Finally, we incorporate a component of the climate rule process.
  Keywords: Morocco; Agriculture; climate change; Green Morocco ,data min- ing Algorithm; data analysis.
  
  
• Automated Greenhouse system for Tomato Crop using Deep Learning  
  by Nagaraj V. Dharwadkar, Vandana R. Harale 
  Abstract: In India, many farmers are facing the major problem of crop diseases. These diseases are affecting growth and the quality of the crop. The crop diseases will occur due to change in the environment variables. To reduce the impact of diseases, the farmers required to continuously monitor the health of the crop. The only solution to this problem is monitor the health of the crop using an automated greenhouse system. In this paper, we have proposed an automated greenhouse system for tomato crop, considering the six climate variables like Temperature, Air Humidity, Soil Moisture, pH Value, CO2, Light Intensity. The proposed system uses Deep Neural Network model for recognition of change in environmental variables. The results shows that the Deep Neural Network (DNN) model is able to reach the accuracy 90% in recognition of change in environment. By monitoring environmental facts, we can able to reduce the impact of diseases and improve the quality of the tomato crop.
  Keywords: Automated Greenhouse System; Climate Variables; DNN Classifier; Tomato Crop; Control System; Crop Diseases.