System implementation of the business analysis of QlikView Business Discovery in NTS companies (Minsk)
Customers: Native land Contractors: A2 Consulting Product: QlikViewProject date: 2012/04 - 2012/09
|
The project team from the customer | Integrator Consultant |
не указана | Potter Alexander (project manager) Gritsenko Aleksandra (Leading developer) Pastushenko Dmitry (Leading developer) Bartosz Alexander (Business analyst)
|
Control of data together with QlikView! The A2 Consulting company completed system implementation of the business analysis (BI) of QlikView Business Discovery in companies SOOO "NTS" (Minsk) - one of the most dynamically developing retail networks In Belarus.
The solution to implement BI-a system in company SOOO "NTS" was caused by the need for the tool which will help with fast creation and visualization of huge data array. In spite of the fact that in the company not the first year is used by SAP, the decision to form analytics based on QlikView was made. It should be noted that legal entities with the different systems are a part of retail retail chain stores under control of SOOO "NTS", and these are more than hundred shopping facilities. Considering the amount of business of the company, process of obtaining the integrated reporting of the existing sources resursoyemok, required special knowledge and took a lot of time. Obtaining difficult reports from the trading systems reduced efficiency of decision making and had negative effect on operational activity of the company in general. Data are stored in the different systems that leads to duplication. For creation of an analytical system it was necessary to work with different data sources.
At creation of the analytics which is based on so different sources there is a problem of "accuracy of the data". According to the known term GIGO - garbage in-garbage out "garbage" which is included in the analytical application turns results of work of analytics into "garbage". Most often it happens because of duplicates of data and incorrectness of primary information. Different names of identical positions and incompleteness of data in reference books lead to distortion of analytical information. In most the enterprises decisions of operational planning are made on the basis of data of the analysis and then the wrong analytical data lead to the wrong solutions. Also mismatch of data in information systems leads to a situation when at everyone "the truth".
Proceeding from heterogeneity of data and impossibility of synchronization of data in all systems, the decision to begin implementation of analytics of QlikView with the Control of Data application was made. The main objective of the application - comparison of records from reference books of different enterprise information systems for the purpose of their reduction to a uniform type. In the project progress data sources and the majority of fields with which future analytical applications will work that reduces time for writing of terms of reference and implementation of the subsequent applications were described. The selected fields from the different accounting systems are compared that allows to trace unambiguously and quickly malformed data, to find duplicates of positions, to check compliance of the reference books given to rules of maintaining. For example, the gross weight should be less than a net weight, or the barcode should be 12 characters, or in a description of goods there have to be no special characters.
Creation the application which daily updates information more than thousands of fields on the following systems was a result of work:
- SAP
- LS-trade
- 1С 8.1 Trade management
- 1С 8.2 Accounting
- Supermagician
Data-refresh happens daily in the automatic mode. It is possible to estimate relevance of data each working day, and to eliminate errors of information input promptly.
In the Control of Data application directory synchronization indicators are calculated that visually displays a situation in general on all sources. Implementation of this application gives understanding, what sync level of data in the different systems, it first of all affects completeness of the analysis.