A DATA QUALITY MODEL OF INFORMATION-SHARING IN A TWO-LEVEL SUPPLY CHAIN

Data quality affects decision quality. In any forecasting and estimation model, the quality of the forecast data or the estimated data needs to be evaluated before using for decision-making. In two-level supply chain model, the retailer collects the sales information and use it to forecast the future demands over the lead-time, and the manufacturer collects the information about retailer's orders and its sales data (if possible) to forecast the future demand over the external supplier's lead-time. However, the quality of the forecast data will decrease and needs to be measured as the time period increases. Hence a quality dimension is needed in this case for indicating the quality level of the estimate. This research defines a quality dimension, “reliability”, and provides a mathematical method to measure the reliability level of the forecast data under two cases: information sharing and non-information sharing [7]. With the outcome of the method, a score ranging from 0 to 1 (including 0 and 1) can be obtained to measure the reliability level associated with the forecast value. This can be a much meaningful indicator for representing the forecast data, thus being helpful for management level to make the right decision.