The eternal paradigm of the forecast
Since unfortunately they are not always known all the requests of customer with the sufficient antelación as to initiate then his manufacture, is necessary the have of a tool that say us that requests go to produce before these same occur. Said tool is the system of forecast.
For the prediction of the requests, splits always of a historical of requests of previous years, and on them effects a mathematical analysis taking into account the estacionalidad, the tendency, the punctual variations, the aleatoriedad and alisamiento and the extrapolation.
Satisfied go trascurriendo said periods, will have to go effecting a new alisamiento of the real data in front of the foreseen initially, that detect and correct the deviations that go producing.
Demand Sensing Like alternative to the forecast of the demand estandard
The forecast of the demand had treated traditionally in technicians and models based in series of time that, like result, predicted a forecast settled in the previous historical sales. The data obtained of several annual series, provided the information on the seasonal patterns that could be predictable.
In the actuality, know that the sales happened are not always the best predictor for the future estimates. It is thus that need distinct technicians that use different indicators to take into account to detect the signals of the demand, between them data from the chain of supply. It is necessary to take into account events of the entrono real in which we manage us, as for example they could be changes in the market, climatic variations, behaviour of the consumer, behaviour of the economy, etc.
treats of a technician of new creation for the generation of forecasts, that combining new mathematical technicians of forecast jointly with the taking of information cuasi real in the chain of supply, allows the create forecasts of high quality, based in the reality of the facts in the chain. The usual error committed by the systems of forecast based in numerical series, remains reduced in a 30 %, or more, with this technician. It is a qualitative jump, that accompanied of an use of concurrent engineering and of the personalización for the customer in sight of the request (MTO), allows to approximate to the reality of the Market, minimising his volatility, obtaining advantages of a chain ‘demand driven', with levels of service to the customer but heights and operations more efficient. This redunda of immediate form in better margins, lower inventory, dragging other indicators of first level to values of excellence.
Up to now are only the big companies with capacities to handle big volumes of data (date mining), that allow the analysis of cientos or thousands of combinations of products, locations and windows of time, have dared with this technician. They attain to convert all this information recopilada, once has been treated with the suitable mathematical models, in signals of demand. The following step is that they are published to effects that the Systems of Planning of the Chain of Supply, can build, distribute and reaprovisionar products, by-products and components of a dynamic and consistent form.
Conscious of the opportunities that this new technician offers, ICIL works in a project of investigation, that in brief will go out to the light and will allow the put to disposal of these technicians to all type of companies, no only the big corporations, so that they can be used with efficiency, clarity and at the same time attain an economic sustainability, minimising as much as possible the impact of the size of the company user.
The investigations that have initiated, are promising and in brief will be able to begin to extrapolate the first data of application in real companies, that have given results very alentadores. In the measure of the possible. Soon we will have the first cases of success.