When data are entered, there can be many different values at many different times. Since the time at which a datum is available is treated as a real number, essentially the same time may appear as different numbers (for example 1991.000 and 1991.001). To try to overcome this problem, the conversion process for data uses a concept called "time slop". Put simply, time slop is the amount of difference in times that are likely to be due to numerical problems rather than true differences in time.
For example, if the time slop is .5 and we have
s1
| 1.2 | 44 |
| 2.3 | 54 |
s2
| 0.8 | 36 |
| 2.0 | 22 |
the values for s1 and s2 are treated as if they occurred at the same times (1.2 and 2.3). This is because .8 is within .5 of 1.2 and 2.0 is within .5 of 2.3. The first two time points (1.2 and 2.3) are treated as new time points, but .8 and 2.0 are never used because they are considered the same as 1.2 and 2.3, which already exist.
You can set a value for time slop in the Global Options dialog, Settings tab. If you set this value to 0.0 (the default), Vensim will use its own algorithm to compute time slop when importing datasets. For most data this algorithm will work quite well. However, if you have a lot of data (with some monthly, some weekly, and some yearly), you will get better results if you set your own value for time slop. Time slop is generally less of an issue when using tabular format data.
