VRP: load constraint exceeded with float digits

[Westlife1002]

cross-posted here: https://groups.google.com/g/or-tools-discuss/c/oqMgzvA0BAU

load (cubic meters) is one of the containts defining the vehicle capacity in my model.

this is how I set this load containt:

def add_cm3_constraints(routing, data, demand_callback_index):
	"""Adds capacity constraint"""
	cm3 = "cm3"
	routing.AddDimensionWithVehicleCapacity(
		demand_callback_index,
		0, # null capacity slack
		cm3_vector(), # vector vehicle_capacity
		True, # start cumul to zero
		cm3)

If I set the vehicle with max load capacity being as "30", the consolidated loads in my vehicle should not exceed "30".
However, I often get the actual load of the vehicle exceeding a little bit, like: 31.45 or 30. 78. The individual loads are float number but I do not think this will affect the max load limit set above. what is the root cause of this error?

UPDATE:
I found "self.assignment.Value(cm3_var)" is actually equal to "math.floor(cm3_var)", hence all the loads (cm3) are floored in the model. why does this happen? does ortools only accept the integer value?

[Mizux]

yes routing solver in C++ use int64_t all python,java, .Net module are "just" wrapper on top of the C++ library with the same limitation.
i.e. our python package is a native python package...