During the population census in Tajikistan using ABBYY FlexiCapture in 7 months 13 million forms were processed

The ABBYY company, software developer and service provider in the field of linguistics, recognitions of documents and data entry, announced in November, 2011 completion of input and processing of forms using the ABBYY FlexiCapture system during the population census and housing stock of the Republic of Tajikistan.

Successful implementing solution of ABBYY allowed to remove need of manual text typing from operators, having left only a task to control correctness of data entry. If earlier, at a manual method, one operator entered 200 forms in day, then when using the ABBYY FlexiCapture system this indicator increased till 1500. As a result input and information processing in the course of the second population census in Tajikistan made 7 months, in comparison with the previous census when the same amount of works at manual entry of data took 18 months. Besides, accuracy of the information increased, and errors became 20 times less, than at a manual method. Now the processed data are stored in electronic form in archive.

The central statistical body of Tajikistan is the Agency according to the statistics at the President of the Republic, responsible for carrying out population censuses, inspections of households, demographic statistics and another.

During the first population census of Tajikistan in 2000 data were processed manually. Such method was ineffective – only on input of all information volume more than one and a half years were required, and the quality of processing left much to be desired.

Considering this experience and planning carrying out the second population census, the Agency according to the statistics considered necessary to use the solution on the automated information input from forms by their scanning and automatic recognition.

Preparation of the project began several years prior to start of a campaign. In 2008 using ABBYY FormReader the trial population census was performed. For the main stage of a census from Tajikistan selected the next version of ABBYY FormReader – the ABBYY FlexiCapture system. This solution is optimal for carrying out such large project as the Republic population census when daily it is necessary to process hundreds of thousands of documents, and dozens of operators are involved in process.

Staff of the Agency and specialists from ABBYY debugged the general data processing system – from registration of portfolios of copyists before creation of output statistical tables. More than 200 jobs for encoders and operators were so prepared, cooperation with the existing infrastructure and databases is established, training of temporary personnel is provided. Special attention was paid to creation of additional algorithms and rules for repeated check of filling of data on forms and their completeness. The second national population census and housing stock of the Republic of Tajikistan takes place in several stages when financing from means of the state budget, with assistance of the World Bank and under the auspices of UN Fund in the field of the population (YuNFPA). From September 21 to September 30, 2010 the bypass of households and filling of forms of a census took place. Further the stage of collecting and preparation of forms for processing followed.

Directly scanning and data entry began in February and completely finished in September, 2011. As a result already at a data entry stage the ABBYY FlexiCapture system allowed to reduce quantity of the errors caused by the wrong filling of forms to a minimum and in need of short terms to find copies of schedules. Especially for this project specialists of ABBYY developed 10 types of forms in two languages – Russian and Tajik. At the moment there passes the stage of statistical processing of results of a census.

"When choosing the contractor we stopped on ABBYY because in this company not only offered us a software package for high-quality data processing of a population census of Tajikistan and shared experience of similar projects, but also provided full technical support of all project, – Bakhtiya Mukhammadiyeva, the director of the Agency of statistics at the President of the Republic of Tajikistan says. – The effect of system implementation was impressive – we managed in the shortest possible time and with the minimum costs to complete processing of 13 million one-sided and bilateral forms, having obtained reliable and complete data".