Abstract - M-Kissan is a mobile phone based agriculture information system to support and assess sustainable agricultural practices. Farmers face a lot of problems for increasing the efficiency of crop production. One of the reasons is that expert information regarding crop cultivation is not reaching farmers in a timely manner. Current e-Governance projects like e- Chaupal works on computer based nodes which are not accessible to most of farmers. Unlike a PC, the power requirement is much less in a mobile handset and a farmer will not have to travel a distance to know the information. Our project M-Kissan is an interactive mobile application which provides all the information needed to implement a sustainable crop plan, helps in field specific nutrient management. The application will answer to farmers’ requirements in their language and send information on issues like weather forecast, pesticide requirement and current commodity prices in village mandis directly to their mobile phones. It helps the farmers to decide from when and how to sow the crop and also when to take their produced crop to market. This information will be available in interactive text format as well as over voice message for illiterate farmers.
Index Terms - Crop Productivity, DSS (Decision Support System), Economic Assessment, e-Chaupal, e-Sagu (Current e-Governance Projects), Software Development, Sustainable Agriculture.
INTRODUCTION
Sustainable agriculture [1] is a term used to refer to a broad group of related practices that emphasize over production techniques that are both environmental sensitive and economically feasible. Indian farmers, however, hardly have any clue of these kinds of agriculture practices. While current e-Governance projects like e-Chaupal, e-Sagu [2] has enough information related to these practices, but it cannot be accessed by a normal villager. This is because there are limited numbers of computer nodes spread over a large area. While these days mobile phone is very common and very easily accessible, even in very remote area, connecting almost each village (and family). Our project leverages the computing power and connectivity of mobile phone to send the agriculture related information. We have developed develop a multi tier mobile application which will be easily accessible and will have rich user interface to give all the information a farmer needs.
PROBLEM DEFINATION
Our problem statement requires us to design and then develop a mobile phone based Crop Information system. Some capabilities of system are that: -
Plan your crop based on region specific soil.
Select a plan recommended by agriculture experts for your soil.
Fertilizer and Pesticide information system.
Sell you crop by placing offers.
Weather forecasts.
Nearest mandi prices.
M-Kissan will be a useful application for the farmers in their agricultural related decisions. It will provide support to the farmers from beginning to the end of a particular crop production. It will be widely used by a larger set of users as it is a mobile phone based application which is very economical and time saving than some of the others current e-governance projects.
LITERATURE STUDY
Computers in agriculture [3]
J2EE/J2ME
Object Relational Mapping
Hibernate Persistence
XML Parsing
JDOM XML API
Screen Scrapping through XQuery
Web Services
JSR-172 mobile client for web services
RSS Feeds Reader
WeatherBug API
Our system will have following parts:-
Mobile Client Application
Server Side Application
Middle Ware Works
Database
Client Server Messaging.
Mobile Client Application
The interactive application developed for JAVA enabled handsets will be menu based and user friendly. It will be developed in J2ME and will have different versions for different kind of cell phones. We has currently made three types of client application.
Simple Java enabled phones.
JSR 172 enabled phones.
Service provider framework dependent (Currently for Reliance R-World).
GSM clients can install the application on their Java enabled handsets and use the features by activating GPRS service.
Since we have exposed web services on our server side, so any vendor can make his own middle layer and provide services based on SMS/other messaging formats also.
Server Side:-
We have made the server side an industry standard enterprise application. We have used J2EE to implement our server side. Server Side Enterprise Application consists of following layers:-
Persistence Layer
Business Logic
RSS Feeds Reader
Screen Scrapper with XQuery
Web Services
Message Formatting
Business Logic does all the business layer processing needed for the application. In our case it is the DTO (Data Transfer Objects) which do the job.
We connect to Weatherbug API to collect weather forecast as RSS feeds so we had to implement an RSS reader to read the weather feeds.
We could not find any RSS feeds for Mandi prices so we used Screen Scrapping to collect the needed data from Agro Market website. It gives us badly formatted HTML pages. we used JTidy to convert badly formatted HTML pages into XMl. Then we made queries in XQuery and used Saxon to fire the queries and collect the needed data.
We exposed web services for all our business logic to make our application open-ended. Other middle layer vendors can use our web services and make their client applications in any technology they want.
Client Server messaging is based on XML formats. In the server side we have used JDOM to format Xml message and on client side we have made parsers to parse these messages. Since the Xml formats makes the whole application very open-ended. This make client and server technology independent. Many mobile phones do not support Java but they support other environments like FlashLite, Brew. So any client application made in these technologies can directly use our server side without any hassles.
FUNCTIONALITIES
Crop Planning
View Plans
Market
Weather
Crop Planning: - The region specific soil data is collected and put into farmer's profile. The on the basis of soil and current season, a list of crops is made. The list of crops contain the information about all the breeds suitable for his soil, the type of seeds he can use price, yield ,manufacturer of the seed, fertilizers available for that crop breed with price ,usage, effects and manufacturer of the fertilizer.
Once the farmer decides which seed type he wants to use, he can set it as his plan. The plan information is sent on server and gets registered on that day.
View Plans: - Farmer can register for many crops in a season. He can view all his plans as a list. He can select a specific plan based on crop name. A crop plan contains following information:-
Current week's information taken from a management timetable for that plan
Fertilizer information for the plan.
Pests and Pesticide information
Disease update of that week
Market: - Farmer can place offers of selling a crop. He selects a crop from the list and input quantity, price and places it. The offer information is sent to server which can be seen by buyers.
Farmers can also view the offers' status. He selects an offer placed by him and gets the information about it expiry, number of buyer who have seen his offer, and phone numbers of 3 buyers who have seen his offer maximum times.
He can also view current mandi prices in his closest mandi. He inputs a commodity and a list of mandis are fetched for him who contains information about maximum price, minimum price and modal price in that mandi.
Weather: - Farmer can see the current weather in his area and a list of next 7 days forecast for his region. Weather information consists of temperature, humidity, winds precipitation.
IMPLEMENTATION
We have used three tier architecture as shown in the picture. The persistence layer connects with database and takes care of all ORM and database queries. We have used simple POJOs with mapping annotations. This makes the application database independent. The business layer uses the POJO's to access data. The relational joins are taken care by annotations and has functions which return the data as a collection of POJOs whenever a join occurs. The fetch type used in persistence is EAGER, which eagerly fetches all POJOs for a join.
Once the data has been fetched from the database business layer uses JDOM to format the data in proper XML formats and sends to the client as a UTF stream.
The exposed web services use the business layer to provide the functionalities.
The weather feeds are read through a simple URL stream. The RSS feeds are parsed and weather data is send to XML formatter which converts it into proper XML message.
The mandi prices are also read through simple URL stream but since mandi prices are returned in a badly formatted HTML message, so we used JTIDY to convert into usable xml and the made a query in Xquery using Saxon to collect the useful data as a string. We send that string to Xml formatter for message formation.
In the client side all the data is received as xml messages, so we have made xml parsers to convert the data into vectors that can be easily use in showing on the screen as lists.
CONCLUSION
Although e-Chaupal and e-Sagu project are there but their penetration to the mass can be questioned. In present world, when the mobile phones are getting very cheap, we can reach to farmers individually using M-Kissan. This will enable the farmers to extract information at their ease.
FUTURE WORK
All the services of M-Kissan can be provided through SMS so a middle layer can be developed. M-Kissan can also be made IVRS compatible so that illiterate farmers can also use it.
ACKNOWLEDGEMENT
We would like to sincerely thank Prof. Ajay Dandekar for his continuous guidance and course correction in our project work It was his motivation and enthusiastic approach towards the project which continuously inspired us throughout our work. His regular guidance and occasional discussions helped a lot to accomplish the task in an efficient manner. We also thank MSc-IT Agri student for sharing their knowledge and providing regular guidance about project.
REFERENCES
Margaret F. Brennan, Pruss Jan A. and Robin G Brumfiels., “A Crop Information Management System to Support and Assess Sustainable Agriculture Practices” Seventh International Conference on Computers in Agriculture , Oct 26-30 1998, Orlando, Florida
B.V.Ratnam, P.Krishna Reddy, and G.S.Reddy, eSagu: An IT based personalized agricultural extension system prototype- -analysis of 51 Farmers' case studies, International Journal of Education and Development using ICT (IJEDICT), Vol.2, No.1(2006).
Pruss, Jan A., Crop Information Management System. Department of Agronomy. The Pennsylvania State University. 1996
Chong-Wei Xu, “A Framework For Developing Wireless Mobile Application”, IEEE Proceeding 0-7695-2613-6/06
P. Krishna Reddy and R.Ankaiah, A framework of information technology-based agriculture information dissemination system to improve crop productivity, Current Science, vol. 88, Num.12 pp. 1905- 1913, June 2005.
Post a Comment