Custom env creation and reward design


#1

I’m looking to create a custom env for my project as I could not find any existing. Environment step (state,action) would be the inventory and price array returned to the agent at discrete time intervals ( say one week) . Trying to get the reward design correct.
1.Is possible to formulate reward in such a way that say if I have 100 units of toy x and I would want to the agent to get the maximum expected return and also sell 85 units before 8 weeks.