From the logistics industry, this problem is definitely harder than it sounds. Scheduling is one thing but scheduling accurately is difficult. I spent two years

From the logistics industry, this problem is definitely harder than it sounds. Scheduling is one thing but scheduling accurately is difficult. I spent two years of my life working on developing a system to consume various inputs and predict when a load would be delivered. This was anywhere from cross country OTR loads to in-town deliveries. The main problem I ran into was the business wasn't willing to limit the inputs to the system and threw everything plus the kitchen sink into the algorithms. This, IMO, unnecessarily complicated the problem. I would give anything to try this system over again.