how so? I guess if u pushed them far enough. I guess what I'm referring to is the MOST binding constraint
Conversation
Replying to
Right - you're talking about the constraints with the highest duals, in optimization terms.
2
Replying to
hmm I don't know optimization. TOC claims optimization has very marginal utility for most throughout problems
1
Replying to
"TOC"? I assume it's b/c optimization is hard to do without an algebraic model - so OR for systems uses simulation models heavily
4
but the same conceptual idea of dual / shadow price applies; there is a gain of X units for relaxing constraint C by one unit.
1
1
If it's a typical optimization problem X is the coefficient of the dual for constraint C
Phrased better: bayt.com/en/specialties
3
1
Replying to
but more fundamentally, he attacks the very concept of per-unit cost accounting. Leads to many wrong incentives
2
1
Replying to
I agree that it can screw up incentives, but attributing costs on per-unit basis can also confuse costs due to business decisions
1
1
Replying to
. Metrics must align with decisions; I'd agree that sometimes shadow prices are a bad tool for deciding which constraint to change
1
Maybe my next post should be on aligning metrics and decisions. But only if I can figure out what to say about it. ;)
2
1


