Negative Actual Setup Hours!?

Hi everyone!
Has anyone ever run into an instance of negative hours in the JobOper.ActSetupHours field? At some point last year, a series of time entries and edits resulted in a value of -10.80 hours in this field for one operation for one employee. I had never seen this before. When you try to view the Job Tracker Details, an error comes up as follows:

An error occurred trying to set a value of '-10.80' in an EpiNumericEditor. The controls EpiBinding is 'JobOper.ActSetupHours'.

The labor entries reflect the correct time for punch-in and out, as well as the right total hours.

I made a BAQ to report out any other instances of negative values in this field. I found a handful more. I was hoping this would shed some light on how this happens.

I don’t know how it happened, and I am not sure how we can fix it. For now, we can fake the data by offsetting another time entry by +10.80 hours, but that really isn’t the right solution.

I am trying to cast a wide net here, anyone ever have negative time issues? Did you figure out how to fix them?

Thanks for reading!

Are they all -10.80 hours?
If so, do the values 648 or 38,880 mean anything (that’s 10.8 hr in minutes and seconds).

Do you know the values should be?

Any odd occurences on those jobs (high scrap, more qty completed than scheduled, job spanned daylight saving time switch, etc…)

No, the values are different for each employee/operation. We looked for those other oddities too, but can’t find any obvious things.
We do know what the values should be, because the in/out times are correct. There wasn’t much of a log of changes to review. I could see that an entry had been edited, but that is normal for us.

To get past the error thrown in Job Tracker Details, try customizing the form and editing the format of that field. If it is in a grid, you might need to do it via code (vs just setting the control’s properties)

I am more looking for the cause and solution, rather than another workaround.

We have a outside system that pushes in Laborhed and LaborDtl records. Sometimes a rare network error doesn’t complete a LaborHed record and the LaborDtl has resulted in negative hours when the shift crosses midnight. It’s a pain to correct as the screen is constantly firing errors about the negative value.

I don’t have an explanation if your workers use MES directly.

Maybe set a BPM on the LaborDtl Update method to throw an error when the proposed burden hours are less than 0. In the error message, write something like “DO NOT CLOSE THIS WINDOW, CONTACT NATE IMMEDIATELY”.

I couldn’t figure this out, so I submitted a support ticket. I will post back if I get a solution.