Agent-written heuristics versus commercial solvers on unsolved job-shop benchmark instances
Authors: Andre Albrecht, Enzo Morosini Frazzon, Leandro C. Coelho, Bruno Vieira
Organizations: ProLogIS – Production and Logistics Intelligent Systems, Department of Production Engineering, Federal University of Santa Catarina (UFSC), Florianópolis, SC, 88040-900, Brazil · CIRRELT and Canada Research Chair in Integrated Logistics, Université Laval, Québec, QC, G1V 0A6, Canada · Department of Industrial Engineering, Federal Institute of Santa Catarina (IFSC), Caçador, SC, 89503-550, Brazil
Commercial solvers are the standard tool for difficult scheduling problems. We study whether a frontier large language model, used as a coding agent with open-source tools only, can match them in quality and speed. We report every run, and independent checkers verify every schedule. On the 100,000-operation job shop tai_100_1000_1, an agent-written solver reached 525,515, below OptalCP, Hexaly, and CP Optimizer. Unchanged, it improved all ten instances of the family and closed two. With equal budgets on the same workstation, it ended 0.0-1.8% above the lower bound, against 2.6-5.8% for CP Optimizer and 2.4-4.4% for Hexaly. On the Behnke-Geiger flexible job shops, an agent-written tabu search improved 11 of 32 open instances. With two transport robots, an agent-written solver improved seven of ten MKT instances. New lower bounds show that four published MKT values cannot be attained on the distributed instance files. Agent-run improvement campaigns closed eight further instances.
Figures & tables
Problem
Instance(s)
Jobs
Machines
Operations
LB / UB
Status
Job shop
tai_100_1000_1
100
1,000
100,000
522,298 / 533,080
open
Flexible job shop
behnke16
100
20
500
391 / 412
open
behnke14
50
20
250
221 / 230
open
behnke31
50
40
250
226 / 226
closed ∗
Flexible job shop, 2 AGVs
MKT01–MKT10
10–20
4–15
55–240
e.g. MKT10: 386.8 / 525
open
21 Chambers–Barnes
10–15
11–18
100–225
see Table 9
7 open
Table 1: Selected instances
Figure 1: Framework of the study
Item
Value
Agent
Claude Code ( Anthropic, 2026a ) , version 2.1, in all sessions
Models of record
claude-opus-5 (Claude Opus 5) and claude-sonnet-5 (Claude Sonnet 5): job shop and flexible job shop, accessed 5–10 August 2026; claude-opus-5-5 (Claude Opus 5.5): transport robots, accessed 25–26 September 2026, and job shop in mode C, 19–20 October 2026; claude-fable-5-1 (Claude Fable 5.1): flexible job shop in mode C, 19–20 October 2026. Public release ( Anthropic, 2026b ) : Claude Sonnet 5, 30 June 2026; Claude Opus 5, 24 July 2026; Claude Fable 5.1, 1 September 2026; Claude Opus 5.5, 22 September 2026. Default decoding settings of Claude Code; the provider does not expose sampling seeds.
Autonomous runs
Harbor task framework ( Harbor Framework, 2026 ) ; one Docker container per run; quota of 4 CPUs (three runs: 8 CPUs) and 4 GB of memory, although the container reported the 32 hardware threads of the host; budget 18,000 s (5 h)
Container software
Python 3.12.3, GCC 13.3.0, NumPy 2.4.6; OR-Tools only if the agent installed it