Qualification report: time profile plot crashes if no observed data is contained

[Yuri05]

Dear OSPSuite experts, I downloaded the files available in:
https://github.com/Open-Systems-Pharmacology/Evaluation-plan-template/tree/master

and executed it with no error. Then, I simplified the 'evaluation_plan.json' file by setting: "GOFMergedPlots": [] and removed the corresponding part in the 'Sections'. This way in the report only 'Concentration-Time Profiles' are plotted (no GoF plots). Executing the reportingEngine worked well. Then, in PK-Sim, I loaded the 'Compound.json' and removed the observed data in the 'Sim 1 - intravenous' simulation and export the updated file as 'Compound.json' and executed the createQualificationReport. However, this time I got this error:

! Warning argument 1 is not a vector
Configuration Plan Information:
Project: Compound
Simulation: Sim 1 - intravenous
SectionReference: ct-profiles
Plot: list(...)
PlotNumber: 0 (json numbering starting at 0); 1 (actual plot index)

Error Trace
↪ createQualificationReport(qualificationRunnerFolder, pkSimPortableFolder)
↪ workflow$runWorkflow()
↪ self[[plotTask]]$runTask(self$configurationPlan)
↪ self$getTaskResults(configurationPlan, self$settings)
↪ ifNotNull(timeProfilePlan$Plot$Type, plotQualificationPopulationTimeProfile(simulationAnalysis = timeProfilePlan$Plot$Analysis,
↪ plotQualificationMeanTimeProfile(configurationPlanCurves = timeProfilePlan$Plot$Curves,
↪ updateQualificationTimeProfilePlotConfiguration(simulatedMetaData = simMetaData,
↪ order(observedMetaData$id)
Error: argument 1 is not a vector
Configuration Plan Information:
Project: Compound
Simulation: Sim 1 - intravenous
SectionReference: ct-profiles
Plot: list(...)
PlotNumber: 0 (json numbering starting at 0); 1 (actual plot index)

here is the zip file:
[Files.zip](https://github.com/GhazalMontaseri/Test2)

Two points:

1. When I create a single datapoint observation at time=0 (or any time) and concentration = 0 mmol/l and add it to the simulation, it works again.
2. population simulations without observation work well, only with individual population I had the issue.

I would appreciate if you can help me with this issue?

Originally posted by @GhazalMontaseri in Open-Systems-Pharmacology/Forum#1569