The aim of this research is to propose a numerical model that treats the variable time so efficient and effective in order to meet the real needs of customers, users and society in general. However, was performed a survey bibliographic on public management, in respect of public works, statistics and operational research to the organizational system, aiming to numerical modeling. The research was based on quantitative methodologies, with emphasis on operational research for the study of public works performed under the management of PCU/UFPA. In developing the database, information was collected construction, renovations and expansions, implemented during the period 2006 to 2009, with the Permanent Commission for Bidding (CPL) and the Foundation that support the Research Development (FADESP). By linear regressions and after the transformed functions were obtained for the model prediction the statistical parameters: correlation coefficient (R) of 0.899, the coefficient of determination (R²) of 0.808, the coefficient of determination (adjusted R²) of 0.796 and error standard (Se) of 0.41. These parameters show a strong linearized correlation between the variables, indicating that 79.60% of the variability of time to execute a public work is caused or produced by variations together the area, the budgeted value, the operational capacity of
IFES; operational capacity of the company; the type of service, and the season. With the results, it was concluded that it is possible to apply and implement the prognostic model for public works, considering that it is a powerful tool in its application to improvement of administrative procedures, both in structure and in its performance, whose main result is forecast variable “time of execution” for the performance of public enterprise.
|
