https://www.amazon.com/Introduction-Materials-Management-Ste...
https://www.amazon.com/gp/aw/d/1260084841/ref=dp_ob_neva_mob...
If you're highly technical and would like to learn how to abstract and solve some of the hardest problems in supply chain management, I would actually look into the field of Operations Research. It is basically mathematical modeling and optimization of real world operational and logistics problems. It's so mathematically oriented that it typically is in the mathematics or science departments in universities, as opposed to management. But it is very much an applied specialization of math, directly applicable to supply chain management and other business/logistics areas.
Interestingly, the bullwhip effect is extremely similar to another problem that a lot of people here have a casual interest in: the traveling wave phenomenon of traffic. The requisite similarities are the same. In order for the pattern to appear, there have to be independent actors with cascading delays in the aquisition of information about the future, and a mismatch in the ability to stop production (decelerate) and start production (accelerate). The severity of those traveling waves may change with changes in the severity of the required conditions, but as long as those conditions exist, the traveling waves will always occur.
Interesting consequences also apply. It is mathematically provable that autonomously controlled vehicles cannot get rid of traveling traffic waves. There will always be a delay in information about what lies ahead, whether it is via the limitations of line of sight, or the latency of radio-based communication. And production cars will always stop faster than they accelerate, because stopping has a safety necessity that doesn't exist for acceleration.
Autonomous cars may have faster reaction times than humans, but at highway speeds reaction times are typically 20% or less of stopping times. Therefore they can slightly mitigate the problem. However, notice how everybody talks about how autonomous cars can travel closer together. What this effectively means is that cars can travel closer to the extent that their reaction times are better. This improves unperterbed throughput slightly, but it would actually make the shock-induced traveling waves worse. All it takes is for a stray plastic bag to wander into the roadway and a car to think it's a dog and brake for it. The cars closest to it stop extremely fast but then accelerate slower than they decelerated. Cascading information delays ensure that the wave continues to travel and increases in severity until it reaches some location where buffer spaces are large enough to absorb the wave (due to lower traffic density).
The supply chain parallel problem emphasizes the importance of inventory. Running super lean on inventory can mean lower costs and higher throughput, but it exacerbates the bullwhip effect substantially.