Search

• The ES capitalizes on the knowledge and experience of specialists in livestock production. • The ES shows 86.67% reliability for the control of livestock practices. • The ES demonstrated that food has the greatest weight in livestock production. This essay documents the development of an "Expert System" based on a Fuzzy Logic model, designed to analyze the outcome a number of variables have on the performance of livestock production (milk and meat) in the Huasteca region of Veracruz in order to support the decision-making of a Sustainable Livestock Production Dynamic System (SLPDS). The Expert System takes into consideration the following input variables: Temperature (T), Rain (RA), Breed (B), Health Plan Implementation (HP), Feeding Plan (FP) and Production System (PS). The aforementioned variables then have an impact on three output variables: Lactation Days (LD), Daily Milk Production (DMP) and Intervals between Births (IBB). Once the Fuzzy Logic model has been developed, an assessment of the variables is made through the Response Surface technique, which allows verifying how the variables behave in the system under study, and their impact on the output variables; as well as, testing different scenarios in order to validate the model and to identify the Livestock Production Systeḿs behavioral patterns. Through the application of Fuzzy Logic regarding the modeling of the 6 variables that impact the performance of livestock production, it is possible to capitalize on the knowledge and experience that producers have and what they have learned based on the observation and practice of many years. Therefore, being able to obtain reliable results that can be shared with agricultural producers and technicians for the improvement of the livestock productivity of the Huasteca region in the state of Veracruz. The Expert System is efficient showing an 86.67% reliability by comparing its results with a panel of specialists in livestock production. The test of the different scenarios shows interesting results when exposing the application of good livestock practices in certain conditions (temperature and rainfall) that maximize the milk and meat production of the region under study. [ABSTRACT FROM AUTHOR]

Highlights • Largest producers of crops include south Asian countries including Pakistan. • There is lack of exploitation of technology for enhancements in agriculture sector. • Diagnosis of crops diseases in rural areas is a challenge. • Intelligent mobile application with easy to use interface has been developed using fuzzy inference system. • Highly promising results leading up to 99% of accurate disease diagnosis have been achieved. Abstract South Asian countries are amongst the largest producers of crops with favourable climate conditions and fertile soil. However, traditional agricultural mechanisms are in place and inadequate effort has been put into exploit the usage of technology. One of the main problems being faced by agriculture sector in Pakistan and other developing countries is that crop diseases are not diagnosed timely and efficiently. Conventional methods for disease diagnosis in crops lead to less accurate and inefficient diagnosis, consequently leading to low productivity. In this paper, an intelligent approach for the diagnosis of crop diseases is proposed which is capable of working over Android mobile devices using fuzzy inference system as the main decision making engine at the backend. The system is capable enough to communicate to the farmers in Pakistan in their local language Urdu and assist them in diagnosing diseases in their crops. Agriculture experts in government sector can get equal benefit from it in diagnosis and prevention of crops diseases. It takes symptoms of the crops as input with a provision of vague input and generates the output in the form of diagnosed disease using its inference engine. The proposed system caters two main crops of Pakistan, cotton and wheat and is capable to diagnose their main diseases. The proposed system has been tested on a pool of 100 real crop problems and its inference engine has shown excellent performance in prediction of the right disease which is up to 99% accurate. [ABSTRACT FROM AUTHOR]

This paper describes an Expert System that can intelligently diagnose diseases in plants. The system is dialog-based and uses a Multi-Criteria Decision Making technique that is a hybrid of Analytic Hierarchy Process and Sensitive Simple Additive Weighting. The paper describes an approach for disease modeling that uses a set of characteristics which are weighted for each disease using two types of weights: Relative Weights and Scales. The diagnostic process involves calculating the utility value for each disease based on the utility values of its characteristics. Experimental results show an accuracy of over 95%. The system implemented is called AgriDiagnose and it consists of a web-based pathology tool to model the diseases and a mobile app for farmers to interact with the system for disease diagnosis in the field. [ABSTRACT FROM AUTHOR]

Production of the Persian lime ( Citrus latifolia tanaka ) has been the main objective of several studies related to the problem of low performance of yield and fruit quality in the orchard, attributed among different technological factors to the minimal application of Good Agricultural Practices (GAP) and to the cultural aspects of the producer. This paper contributes to the recognition of the behavior patterns of GAP for seasonal orchard (SO), to allow the Persian lime producers to make the right decisions assessing and improving the management of their orchards. To identify the behavior patterns in the Persian lime production process an expert system (ES) based on fuzzy logic proposed by Fernández et al. (2014) has been used, in which a set of inference rules based on the knowledge of experts in this field is encoded to explain the interrelationship of the agricultural practices and uncertainties in the production of Persian lime: Pruning, Soil nutrition, Pests Control, Planting density, Tree production, Wind, Rainfall. The ES simulates from agricultural practices and uncertainties, the Persian lime production system in three stages of fruit growth, which represent the fuzzy models of the ES: flowering, bud, and fruit. The manipulation of the agricultural practices in the ES allowed to model production scenarios for SO of Persian lime, and helped to identify behavior patterns in these practices with production yield and fruit quality. The results demonstrate that if prior to fertilization, the practice of “pruning” the tree is performed, orchard productivity increases. However, when the “pruning” (aesthetics or stressful) is performed less than 50 mm month −1 of rain, even in optimal conditions of application of nutrients and pest control, the production yield is similar. The modeling scenarios of the ES provide information regarding behavior patterns to the producer, and the interrelation of agricultural practices in uncertain environments of rain and wind in order to improve the decision-making process in Persian lime production. [ABSTRACT FROM AUTHOR]

Highlights: [•] Several machine learning algorithms for the classification of yield components were compared. [•] Rules to infer yields at harvest of durum wheat with data from Buenos Aires Province were found it. [•] A priori method obtains the best performance for all locations. [•] More relevant traits for yield levels prediction were identified. [•] Thousand grain weight is the most relevant trait for predicting durum wheat yield. [Copyright &y& Elsevier]

Abstract: Agricultural land suitability evaluation for crop production is a process that requires specialized geo-environmental information and the expertise of a computer scientist to analyze and interpret the information. This paper presents ALSE, an intelligent system for assessing land suitability for different types of crops in tropical and subtropical regions (e.g. mango, banana, papaya, citrus, and guava) based on geo-environmental factors that automates the process of evaluation and illustrates the results on an attribute table. Its main features include support of GIS capabilities on the digital map of an area with the FAO-SYS framework model with some necessary modifications to suit the local environmental conditions for land evaluation, and the support of expert knowledge through on spatial tools to derive criteria weights with their relative importance. A dynamic program for calculation of eigenvalues and eigenvectors of a weighting matrix is provided. Expertise and knowledge help ensure that ALSE databases represent realistic, practicable and functional systems. It is useful for decision makers to determine the quality of land for agricultural uses and is intended as a decision and planning support. Responsibility for any decisions based partly or wholly on the output of ALSE rests with the decision maker. ALSE ensures that the results are interpreted correctly within the relevant context, and contributes by maximizing land-use planning and decision support. [Copyright &y& Elsevier]

Abstract: One of the essential requirements of current agricultural practice is the need to evolve towards crop techniques that make better use of productive factors behaving in sustainable and environmentally respectful ways. Pest control is one of the most important problems to take into account, due to the significant production losses that pests may cause. The Integrated Production for olive crop defines a set of rules that has to be followed in order to ensure a production of higher quality, and promotes an environmentally respectful model of Olive cultivation. This work presents SAIFA (spanish acronym for Sistema de Alerta e Información Fitosanitaria Andaluz – Andalusian Phitosanitary Information and Alert System) a web-based information system which allows monitoring the Integrated Production for the olive crop in Andalusia, Spain. SAIFA has been built as a tool to assist agricultural technicians to comply with the quality standards of Integrated Production, and to help decision makers to choose the actions to be performed in the crop. Also, it assists coordinators to choose an Integrated Production strategy applicable for the whole region with the main objective of ensuring the safety of crops and reporting the current phitosanitary state to the authorities. [Copyright &y& Elsevier]

Abstract: Hops (Humulus lupulus L.) essentially contribute to the quality of the taste of beer and its flavor. Experimental hop breeding data from the Slovenian Institute of Hop Research and Brewing were used to create a model of an expert system based on the multi-attribute decision modeling methodology DEX. The DEX-HOP 1.0 model assessed and ranked individual hop hybrids’ and hop varieties’ breeding potential to be assessed and ranked. The model has 18 attributes, hierarchically grouped within four main attributes: Biology, Chemistry, Morphology and Brewing value. Furthermore, utility functions in the model were defined by sets of elementary decision rules throughout the entire hierarchy for all aggregated attributes. The central part of the model contains 144 decision rules, which were specified according to the model users’ previous breeding experiences. Four Slovenian hop hybrids, A1/54, A2/104, A3/112 and A4/122, and a reference hop variety Hallertauer Magnum with the target characteristics of plant resistance and brewing value were used for the model assessment. Based on the breeding experiences and model results, overall and individual attributes’ assessments were carried out. A3/112 and A4/122 reached the overall level of the reference variety and were thus assessed as appropriate for further breeding. In contrast, the attributes of A1/54 and A2/104 did not meet expectations in relation to the reference variety. Decisions based on the DEX-HOP 1.0 model assessment offered an additional tool for experts’ final decisions in selecting appropriate materials for further breeding or the commercial use of hop plants. Test results validate the application of the model for further research. [ABSTRACT FROM AUTHOR]

Abstract: Animal waste has always been considered as a resource for agricultural input as biofertilizer. However, the management is becoming more stringent due to environmental regulations. Livestock producers are faced with different manure management options that may be implemented into their operations. Given the expansion of the livestock industry, the implementation of environmental regulations, and the increasing importance of social and health issues, the selection of optimal manure management systems is becoming a strategically important task. Increasingly, integrated decision support systems (DSSs) are becoming necessary to assist decision makers in their evaluation of different manure management alternatives, like, liquid system, semi-solid system, solid system and bio-gas or bio-energy system based on combinations of different manure management sub-systems (collection, storage and application). To address this situation, a user-friendly computer program called Integrated Swine Manure Management (ISMM) is being developed for the Canadian Prairie provinces. Decision criteria including environmental, agronomic, social and health, greenhouse gas emission, and economic factors have been considered for the selection, design, and operation of the DSS. The expert system modeling is based on Visual Basic programming. Decision on adopting a particular combination of systems components is based on performance rating of the overall system. The program is interactive so that weighting factors for the different decision criteria can be varied to suit site-specific considerations. In this paper, the systems approach for development of an integrated liquid manure management system is discussed. Using a case study, sensitivity analysis of different combinations of management components is also reported for systems performance. The decision software compared satisfactorily with other available DSS packages. [Copyright &y& Elsevier]

Abstract: Location awareness is essential for many Precision Farming (PF) tasks with strong spatiotemporal, environmental, public health and food safety characteristics. Nevertheless, its role is much more crucial in PF tasks with efficacy depending mainly on local climate conditions and the collaboration of users. A PF task with the aforementioned characteristics is the insecticide-bait ground spraying against olive fruit fly, the most serious pest on olive cultivations. It requires location awareness, so as to be more efficient, friendly for the environment and the domestic areas, and ensure olive products with low insecticide residues. This research proposes an innovative, integrated, Location-Aware System (LAS) suitable for the ground control of the olive fruit fly. The developed system enables rapid prototyping of Location-Aware (LA) services in an intelligent PF environment combining location sensing technologies with wireless Internet, Geographical Information Systems (GIS), and Expert Systems (ES). We focus on the functional and operational capabilities of the middleware architecture, on the design issues of the developed GIS, ES, and LA modules, as well as, on the factors and infrastructure that must be considered during the spraying process. Based on this framework we developed specific LA services, such as finding the area to be sprayed, estimating the amount of the spraying solution required, canceling the spraying process, etc. These services aim in a more efficient and environmental friendly treatment. To validate the LAS a moderate-scale experiment is performed showing that the proposed system is functional and operational. LAS consult effectively the tractor attendants on how to spray, by means of reducing spraying failures and minimizing the decisions that must be taken during spraying process. Preliminary results report that with LAS no over sprayings occur, sprayings are based on infestation risk, cultivation characteristics, and meteorological conditions. Finally, a safe distance from biological cultivations, environmental protected and domestic areas is kept, avoiding pollution of these areas with insecticide residues. [Copyright &y& Elsevier]

Abstract: This paper presents Coptimizer, a model-driven decision support system designed to help growers to optimize and track the use of copper-based fungicides against grapevine downy mildew in European organic viticulture, where the use of only a fixed amount of copper per year per hectare is permitted. Copper is a preventive fungicide allowed in organic agriculture that is active only at the application site (i.e. it is non-systemic), so that new plant growth results in unprotected tissues. In areas where disease incidence is high, organic farmers usually apply weekly copper treatments at the risk of exceeding the fixed threshold. The capability of Coptimizer for not exceeding the copper threshold was evaluated in simulation using historical data and tested in two experiments under field conditions during 2008. The results show that by using Coptimizer growers could be able to maintain the same level of protection as that gained by traditional application schedule while applying only half the amount of the fungicide. [Copyright &y& Elsevier]

Abstract: A web-based expert system was developed to help potato growers in Michigan make important decisions on optimal use and timing of fungicide applications to mitigate the risk of potato late blight (PLB) disease development. The system was developed around a relational database which stores information on PLB risk which is generated using a PLB prediction model implemented in Perl. The model uses data on relative humidity and temperature from a network of automated weather stations in Michigan to estimate if environmental conditions are favorable for PLB epidemic development. Management recommendations are then produced based on the level of risk. Information stored in the database is presented to users through a series of interactive web pages which are dynamically generated through a combination of Perl and PHP scripts which interface between the website and relational database. This system differs from traditional systems in two major ways. Firstly, the system is web-based and interactive, allowing growers to access site specific, weather-based recommendations through their personal computer without the purchase or installation of additional equipment and/or software. Secondly, the system incorporates geographic proximity of known PLB occurrences and data from a network of regional weather stations into daily risk estimates and management recommendations for 48 locations in Michigan. Recommendations include application timing and spray rates for common fungicides. The interactive website is accessible over the web at URL: http://www.lateblight.org. [Copyright &y& Elsevier]

A web-based system was developed to advise on the relative efficacy of different herbicides for mixes of weed and crop species at different times of the year in a forestry or farm forestry setting. The system assumes that weed identification and impact assessment or prediction have already been accomplished and that there are no cost-effective non-chemical alternatives. The expert system produces a relative suitability index for each herbicide, as well as an English language discussion of the case. [Copyright &y& Elsevier]

This paper presents a methodology for developing a strategic Expert System for wheat crop production, which has been developed using a Multiple Design approach. This Expert System consists of six subsystems; namely, variety selection, land preparation, sowing, irrigation, fertilization, and harvest. These subsystems exchange data between them through a common database. Each subsystem advises wheat growers on the corresponding plan(s), comprising a set of agricultural/chemical operations before or during the growing season. Applying an appropriate plan reduces risk to farmers from having a problem during the growing season. The system was verified, validated, and, tested in the field. Beneficial economic and environmental impacts have been demonstrated. [Copyright &y& Elsevier]