A measurement through the tools provides an FSR of 18 which is real close to our simulated results.  This was examined and detected graphically as there was issues trying to connect our calibration and the envelope data into the tool.
The group index was calculated for other prior examples with the toolsets but due to tool errors it was bypassed for MZI4 at this time.  However an estimate then was given 4.19 based off of simulation.
Given this value our group index would lie within the corner analysis shown above.

Conclusion

The MZI design and simulation given performing those simulation for fabrication errors can reliably provide expectations for circuit fabrication.  There is error within fabrication and specifically the imperfections and limitations for the gradient couplers to get light in and out are major restrictions when designing and understanding optical silicon photonics capabilities for performance compared to ideal situations.  The reason for the frequency shift to higher wavelengths is assumed to be due to the center frequency of the laser that was tuned to perform the calculations.  Ideally thing would have been centered at 1550 but we can see a 30 nm positive shift in the middle power of spectrum during our tests.  This was the most alarming issue.
Due to analysis being performed following matlab license expiration, and leveraging python code that needed modifications it was determined more investigation and correct scripting, tool setups are needed to derive the exact measured group index.  However given simulated estimates and FSR as seen in measured results within the correct range, it was calculated to put a similar variance on the group index which still brought it very close within simulation expectations.