WEB TECHNOLOGIES USED FOR THE PROGNOSIS OF THE QUANTITIES OF WASTE PRODUCED
Abstract
In this article the authors propose a modality of prognosis of the quantities of waste generated in a certain period. The proposition was finalized by achieving a model of prognosis present on a site hosted by a Web server. Time series with three components were used: trend, seasonality and residual variable. According to the input data the adjustment model regarding the description of the analyzed phenomenon (additive and multiplying) is chosen. If the chronogram of the analyzed phenomenon indicates an oscillating evolution, of sinusoidal nature, the additive model (+) is chosen, otherwise the multiplying model ( ) is chosen. The seasonal component is estimated and the deseasonalized chronological series is determined. The seasonality is determined by: the procedure of arithmetical means, the procedure of moving averages and the procedure of analytical tendency. The adjustment function is specified regarding the tendency of the phenomenon and its parameters are estimated. The significance of the adjustment function is verified specifying the significance threshold with which it may be accepted as significant. The estimation of the parameters of the adjustment function is made on the basis of the application of the method of least squares. The values of the phenomenon on two or three trimesters/months are estimated for the processed data (by using time series with two or three steps ahead). Several types of waste specified by the value of the meter for waste from the database may be accepted. The model proposed makes decisions and justifies if the data prognosticated is accepted or not. In the study the following tests are used: F test (Fisher – Snedecor), student concordance test (t) (for testing the significance of the results). All information from the input data to the final data is stored in a MySQL database.
Full Text: PDF
![Licenţa Creative Commons](https://i.creativecommons.org/l/by-sa/4.0/88x31.png)
The articles in this journal are licensed under a Creative Commons Attribution-ShareAlike 4.0 International License .
Since 2008. Maintained by Livius.