1. Error Message Problem, reported by NVEnergy (March 28, 2011): If running only the Policy Model, not Simulation and Policy Models, and if the disruption months exceed maximum disruption months, the error message is incorrect. The message is Error Number 1000. It should be Error Number 1536. Error 1536 has a detailed description of the reason for the error message.
This is not a bug. The problem occurs when the file ERRMSG.TXT is missing from the program folder. The file is placed in the program folder at installation.
2. Delete Disruptions Problem, reported by Steve Chapel, October 2010: If you delete all disruptions and exit UFIM the next time the program is opened the user interface will not work -- the model expects at least one disruption to be in the data set. You must have a disruption even if it is not used in your analysis.
This is serious bug! If you delete all disruption and exit the model - this will make your data set invalid and the only known solution is to rebuild the data set from scratch, or from a backup copy
This bug has been fixed in UFIM Version 5.01
3. Replacement Power Cost Curve Problem, reported by Steve Chapel, March 2011: If you select the option to enter replacement power costs based on the absolute number of megawatt-hours (in the Basic Data input screen) the input screens for entering shortage cost data will not work.
This is not a serious problem because the input data can always be formulated to use the alternative option - replacement power cost based on fractional burn reduction.
4. Exporting Data to Spreadsheet Problem, reported by Duke Power, April 2012 and verified by Steve Chapel: If you attempt to export output to a spreadsheet it only works for the Summary data. For all of the other files (Normal, Disruption, etc) the model will crash. The problem may also be causing errors with the disruption reporting when maximum inventory is set to 80.
Chapel determined that this occurs in the C program that generates the output files. This was a serious bug and took significant time to diagnose and repair. The problem has been fixed and the solution is available. The solution requires that you replace the UFIMReport.dll file with a new file from Chapel.
This bug has been fixed in UFIM V5.01.
5. Converting Non-Seasonal Case to Seasonal Case Problem, reported by Duke Power and verified by Chapel, August 2012. I did some testing and indeed there is a bug in the user interface. If you create a working non-seasonal case and then convert it to seasonal, the case will not run (you get an error 1000 – Invalid use of NULL). The work around is to create a new case, define it as seasonal on the Basic Data tab and then reenter the case data.
I will fix this bug as soon as I sort out why it is happening. This should not be a serious issue because there is a workaround and most people are running seasonal cases.
6. Editing Demand Curves and Burn Reduction Cost Curves, discovered by Southern Company and verified by Chapel, October 30, 2013.
On Windows 7 if you change the display from the default of 100% to the 125% or 150% options the scroll box that lists the curves does not work. The solution is to set the display value back to the 100% default option.
At this time there is no plan to change the code to fix this bug.
7. Editing Normal Times Supply Data for Nonseasonal Case, discovered by Chapel, November 20, 2014.
When entering normal times supply data, if you do the following series of changes and model runs you will get an error message that the input data is not valid: (1) select the option that the average cost of fuel depends on the amount ordered and enter a supply curve and run the model, (2) select the option that average fuel cost is independent of the quantity ordered (non supply curve) and run the model, (3) enter a supply curve and attempt to run the model.
This is a bug in the interface that only occurs for the nonseasonal option. There is a work around: at step (3) after entering the supply curve if you go to the Basic Data tab and select the Seasonal option and then select the Nonseasonal option and save the data, the model will run correctly.