This is by far one of the best logistics articles I’ve ever read and the data visualization paints the perfect picture. I wonder what it looks if you can run algorithms with a potential pickup service or if the shoppers begin to drop off the groceries inside the homes? Or at a refrigerated drop box located in a central area or at a lobby where you have at least X amount of occupants using the service? Would be interesting to see as one would think this would increase inefficiency by opening up the delivery window and adding value for the customer but opens up a brand new set of unanswered questions. Or maybe you already have them answered!