Three questions were defined and posted on the white board in the ED conference room: 1. How should we define the quality of whatever forecasting mechanism we come up with? 2. How well can we forecast patient volume just using historical ED data? In particular, how can we use past patient volume information in the middle of the preceding month(s) to make daily forecasts for the coming month 3. How much could we improve our forecast quality by adding information such as weather or current flu activity