In order for a distribution management system to be predictive of the type described above it must fulfill two important criteria. Firstly the production level or total quality score required by the customers and producers in the chain of distribution must meet or exceed the maximum predicted value from the statistical analysis. Secondly the inventory level or absolute capacity must be equal to or greater than the maximum predicted value from the statistical analysis.
In order to meet these two criteria the production data and/or total inventory should have been analyzed using statistical data. When analyzing the data it is necessary to meet some statistical criterion that will measure the reliability of the sample. The most common statistical tests used are the chi-square, t-test, one way analysis, and multivariate analysis. These various tests can be performed with descriptive statistics and/or ordinal statistical analysis. When using descriptive statistics the degrees of correlation of the variables in the sample will provide the information needed to determine the statistical significance of the variable.
When predicting future job performance it is widely used by manufacturing companies to make the decisions about new product development, new hires and other job openings. The predictive validity of the PV process is used to improve decision making when making important business decisions. In addition to predicting future performance companies use the PV to improve productivity, reduce costs and streamline operations. This is done by reducing bottlenecks in production, eliminating waste, streamlining internal processes and controlling labor costs. The key performance indicators used to measure PV are the job production index (JPI), measure of overall customer satisfaction, measure of new orders and measure of cost reductions. While there are many methods that are being used to evaluate Predictive Validity it is important that it meets the required criteria.