معماری نرم افزار برای سیستم های اطلاعاتی مدیریت مزرعه در کشاورزی با دقت بالا

Farm management information systems (FMIS) have steadily increased in their level of sophistication as they have included new technologies with Internet connectivity being the latest addition. However, few FMIS have used the full capabilities of the Internet, and the emerging concept of precision agriculture has little or no support in the current commercially available FMIS. FMIS for precision agriculture have certain additional requirements to traditional FMIS, which makes the implementation of these systems technically more complicated in several aspects. Our research aimed to identify the requirements posed by precision agriculture on FMIS and then evaluate a modern Web-based approach to the implementation of an FMIS that fulfilled these additional requirements.

1.1. Background information An information management system is required to unite all the parts of precision agriculture as described by Stafford (2000). This system has to be able to store the sensor data and operation documents generated by farming implements during their operation. The same system is also used either directly or indirectly (using an external service) to create the operation plans implemented with the ISOBUS (ISO 11783) enabled tractor-implement combination. For the generation of these plans, not only are considerable amounts of data required, but to create an accurate plan, working biological models, such as those described in Kropff et al. (2001) are required. The information system must be able to manage various data formats, both standardised as well as proprietary and be able to exchange data with services that provide computation for precision agriculture. Besides the features required by precision agriculture, the information system must also act as a traditional FMIS and provide the same record keeping and planning functions as traditional agricultural information systems. Our research aims to identify the requirements of an FMIS for precision agriculture and then proposes a software architecture that fulfills these requirements. This architecture is to be based on the popular Web Application approach1 and is evaluated together with a partial proof-of-concept implementation that was used in a practical demonstration involving a chosen field operation of precision agriculture. The importance of information technology and communication in precision agriculture is well established in research with studies such as Munack and Speckmann (2001) and Santana et al. (2007) supporting it. 1.2. Outline of this article This article is organised so that following this introductory section there is a brief review of related research into modern FMIS and other information systems in agriculture. The review is followed by a set of identified requirements that precision agriculture poses for FMIS, which are then used as the foundation for specifying a software architecture for an FMIS for precision agriculture. Section 5 describes a test case for the specified architecture, which is then used together with a partial implementation to evaluate the architecture.

This article has studied the information systems of precision agriculture. The first section of this article identified the stakeholders, their requirements and the requirements that precision agriculture has for these information systems. As research results, a description of a software architecture suitable for an FMIS for precision agriculture was presented that meets all of the identified requirements. The presented architecture differs from existing FMIS by fully utilising the currently popular Web application architecture and by providing the features necessary for the implementation of precision agriculture. The requirements for an FMIS for precision agriculture are an addition to the requirements of existing commercial FMIS. The new requirements that precision agriculture introduces to FMIS are those of increased connectivity and the management of GIS data. The increased connectivity of the FMIS is communication with the external services of precision agriculture and communication with the ISO 11783 tractor-implement combination that carries out the field operations. The management of GIS data is a fundamental requirement of precision agriculture and is considerably more information-intensive than the information management of traditional agriculture. The chosen Web application architecture would appear to provide the optimal combination of features, ease-of-use and availability compared to the software architectures for on-site software. Evaluation of the system suggests that such a system would be convenient to use and has considerable inherent extensibility. Furthermore, the evaluation of the system, and the partial proof-of-concept implementation of the described system, suggests that the information systems of precision agriculture can be implemented with technologies currently available today. The Web application architecture can be criticised for its user interfaces, the strong coupling between the availability of the system with that of an Internet connection and the complete off-site storage of data. Indeed, great care must be taken in the design of user interfaces as poor interfaces have been identified as a probable cause for low adoption of information systems in agriculture. However, Web interfaces are not inherently inferior to those of on-site software and can, with proper design, be made as functional and as usable as those of on-site software. Complete off-site storage of farm data is in the best interest of the farmers as their data is considerably more secure on a well-backed-up central system than on a volatile local farm PC. The off-site storage of data is likely to raise issues regarding privacy and while some of these issues can be expected to be somewhat irrational in nature, they are still critical for the uptake and acceptance of any system. While the technical aspect of privacy is often easy to achieve with proper authentication, ensuring the user base of this can require considerable efforts in salesmanship. Currently, precision agriculture has yet to reach the bulk of plant production as most of the existing machinery and information systems are insufficient for the utilisation of precision agriculture. However, since the features required by precision agriculture are becoming more common in manufactured hardware, the natural phasing out of old equipment eventually makes the large-scale utilisation of precision agriculture feasible in the near future; effectively generating the need and market for information systems with support for precision agriculture.