REVISTA DYNA NEW TECHNOLOGIES

Key words:

industry application, automatisation, business forecast financial statement, software system, neural network applications, software system

Article type:

ARTICULO DE INVESTIGACION / RESEARCH ARTICLE

Section:

RESEARCH ARTICLES

ABSTRACT:
The establishment of sales targets and monitorization of the business forecast financial compliance in large IT multinational companies are very complex processes. Both require Access to the corporate systems to extract and to manage stored data. This is carried out in order to handle the accounting records registered in the financial system relating all this information with one another. These processes, where a large number of actors are involved (local, regional and global organizations, managers and technicians), are very stressful for them and extremely prone to errors. Reports should be generated quickly by the technicians so that the senior manager can take important decisions with respect to the company. This research aims to show that an approach based on Neural Networks (NN) can improve the aforementioned processes. We have implemented a solution based on NN and software agent (SA) that optimized these processes in a large IT multinational company. The objectives were to establish a realistic sales targets for each product and comercial area (at local, regional and global levels), to achieve a reduction of more than 45 per cent in the time consumed executing the processes, automatizing part of their tasks and eliminating human errors.
The implemented solution automatically relates the estimated values in each quarter for order books (OB), net sales (NS) and unadjustment margin (UM) of projects, programs and contracts included in each deal with the information available in the corporate systems and in the financial system of the company. Historical information was also used to estimate realistic annual sales targets. An average reduction of 99.97% in the periodic monitoring and in the calculation of sales targets was achieved, many stressful activities were automatically executed, human errors were eliminated, and, reports for the managers could be quickly provided at any time. This research shows that the use of artificial NN can improve the processes of sales goal-settings and the supervision of business forecast financial compliance in a large multinational IT company.