No abstract

Forecasting is one of the logistics activities and a sales forecast is the starting point for the elaboration of business plans. Forecast accuracy affects the business outcomes and ultimately may significantly affect the economic stability of the company. The accuracy of the prediction depends on the suitability of the use of forecasting methods, experience, quality of input data, time period and other factors. The input data are usually not deterministic but they are often of random nature. They are affected by uncertainties of the market environment, and many other factors. Taking into account the input data uncertainty, the forecast error can by reduced. This article deals with the use of the software tool for incorporating data uncertainty into forecasting. Proposals are presented of a forecasting approach and simulation of the impact of uncertain input parameters to the target forecasted value by this case study model. The statistical analysis and risk analysis of the forecast results is carried out including sensitivity analysis and variables impact analysis.

No abstract

Demand forecasting is very often used in production planning, especially, when a manufacturer needs in a longer production cycle to respond flexibly to market demands. Production based on longer-term forecasts means bearing the risk of forecast unreliability in the form of finished product inventory deficit or excess. The use of computer simulation allows us to improve the planning process and optimise the plan for the intended goal. This paper presents the use of quantitative forecasting and computer simulations to create the production plan. Two approaches to production plan creation are demonstrated in a model case study. Products are characterized by varying demand and are produced on a single production line in continuous operation. The first approach uses ARIMA(2,0,2) (Auto-Regressive Integrated Moving Average) prognostic method selected as the most reliable method based on MAPE (Mean Absolute Percent Error). The second method applies Monte Carlo simulations and optimisation. The aim of the plan optimisation is minimisation the total costs connected with line rebuilding and storage of products. The comparison of the two approaches shows that planning using computer simulations and optimisation leads to lower total costs.

No abstract