A quiet tape: no new papers, and the practice news is mostly vendor claims
The day in optimization
No new items came in today, so there is no theory to report. The practice side was also thin: searches turned up no deployment news dated this week. The most useful item is a forestry routing study, which includes a field comparison rather than only a vendor's claim.
Theory
Nothing new today.
In practice
A new arXiv paper covers real-world, large-scale, multi-period log truck routing and scheduling in Canadian forestry. It reports a 35% cut in total distance travelled. The authors compared the tool against current practice over 50 weeks. The partner's management estimated the annual savings, so they are not independently audited. The optimizer now sits inside a web-based decision-support system that dispatchers can use. That deployment detail matters more than the headline percentage. Routing tools only help if dispatchers actually use them.
Most other routing "success stories" are vendor material and should be read with care. One example is Sphere's case study. Sphere reports that a North American logistics operator cut fuel costs by 17% and improved on-time delivery by 15%. It also says the investment paid back within eight weeks. These are the vendor's claims, and the case study does not describe how they were measured.
Worth watching
- A vendor-run comparison from Hexaly claims that OR-Tools, Gurobi and CPLEX fail to find feasible solutions within 6 hours on very large job-shop instances. This is a competitor's benchmark, so treat it as a claim until others reproduce it.
- PeerSpot's solver reviews page lists Gurobi at 34.8% mindshare, down from 36.9% a year earlier, with CPLEX at 25.1%. This is a user-survey measure, not a measure of market share.
Try ProvenDecisions free: type, photograph or say an optimization problem; every answer is checked by a solver.