Forecasting model analysis for aesthetic component production in fashion market

The fashion market involves worth trillion of USD all over the world, and just in Italy worth about fifty billion euros. In Europe the market is divided in supplier countries and important brand headquarters country. Italy is one of the most important supplier countries with highest employment and export level. It is dominated by small and medium enterprises which have enjoyed a rapid growth. These SMEs have now production management and monitoring issues. The thesis has been carried out in Goretti S.r.l. company, one of these SMEs Goretti S.r.l. implement time relief to control production cost with latent purpose of creating reliable quotes, then not subjected to uncontrollable variables. The system currently in place is unreliable (it may have considerable shifts in production phase), it does not allow the creation of a historic data storage with a strong statistical basis and therefore it does not permit the creation of a forecasting model. It is necessary to purify the data from endogenous and exogenous factors that hinder their interpretation. To create a forecasting model first step is to elaborate a solid statistical data storage. A benchmark technology has been selected and based on this a data collection started using Bedaux chronotechnical technique. After the collection of considerable amount of data it was run the analysis of the database to select proper data. Data were reworked with mathematical and statistical instrument to extract relations among them. A model was created to forecast the machining time based on parametric input data obtainable from software used. Finally the model was tested and verified with empirical method on real machining process.