extract: 2023-00-00-sciencedirect-flexible-job-shop-scheduling-review

Pentagon-Agent: Ganymede <F99EBFA6-547B-4096-BEEA-1D59C3E4028A>

domains/internet-finance/general-job-shop-scheduling-is-np-complete-for-more-than-two-machines.md

@@ -0,0 +1,41 @@
+---
+type: claim
+domain: internet-finance
+description: "Computational complexity theory establishes that optimal job-shop scheduling becomes intractable at scale beyond trivial cases"
+confidence: proven
+source: "ScienceDirect review article on Flexible Job Shop Scheduling Problem, 2023; established operations research result"
+created: 2026-03-11
+---
+
+# General job-shop scheduling is NP-complete for more than two machines
+
+The classical Job Shop Scheduling Problem (JSSP) is NP-complete for m > 2 machines, meaning no polynomial-time algorithm exists to find optimal solutions for non-trivial instances. This is a foundational result in operations research and computational complexity theory.
+
+This matters because it establishes the computational boundary between tractable and intractable scheduling problems. When designing coordination systems (like Teleo's pipeline architecture), understanding which side of this boundary your problem falls on determines whether you need heuristics or can use exact optimization.
+
+## Evidence
+
+The ScienceDirect review states: "Classical Job Shop Scheduling Problem (JSSP): n jobs, m machines, fixed operation-to-machine mapping, NP-complete for m > 2."
+
+This is a well-established result in operations research. The proof shows that even with fixed operation-to-machine mappings, finding the optimal schedule that minimizes makespan (total completion time) requires exponential time in the worst case once you have three or more machines.
+
+The Flexible JSSP (FJSP) adds machine assignment as a decision variable on top of sequencing, making it strictly harder than classical JSSP.
+
+## Implications
+
+For any multi-stage coordination system:
+1. If your problem maps to general JSSP with >2 stages, you cannot guarantee optimal solutions at scale
+2. Heuristics and approximation algorithms become necessary
+3. Problem structure matters — special cases (like flow-shop or hybrid flow-shop) can be easier
+4. The choice of coordination mechanism should account for computational tractability
+
+This is why [[hybrid-flow-shop-scheduling-with-simple-dispatching-rules-performs-within-5-10-percent-of-optimal-for-homogeneous-workers]] matters — it identifies a tractable special case that applies to pipeline architectures.
+
+---
+
+Relevant Notes:
+- [[hybrid-flow-shop-scheduling-with-simple-dispatching-rules-performs-within-5-10-percent-of-optimal-for-homogeneous-workers]]
+- domains/internet-finance/_map
+
+Topics:
+- domains/internet-finance/_map

domains/internet-finance/hybrid-flow-shop-scheduling-with-simple-dispatching-rules-performs-within-5-10-percent-of-optimal-for-homogeneous-workers.md

@@ -0,0 +1,49 @@
+---
+type: claim
+domain: internet-finance
+description: "Operations research shows simple priority rules suffice for pipeline architectures with sequential stages and uniform worker capability"
+confidence: likely
+source: "ScienceDirect review article on Flexible Job Shop Scheduling Problem, 2023"
+created: 2026-03-11
+---
+
+# Hybrid flow-shop scheduling with simple dispatching rules performs within 5-10 percent of optimal for homogeneous workers
+
+For pipeline architectures where all work flows through the same sequence of stages (hybrid flow-shop), and workers within each stage have similar capabilities, simple priority dispatching rules like shortest-job-first or FIFO within priority classes achieve near-optimal performance without requiring complex metaheuristic optimization.
+
+This matters for Teleo's pipeline architecture (research → extract → eval) because it means we don't need sophisticated scheduling algorithms. The computational complexity that makes general Job Shop Scheduling Problems NP-hard doesn't apply when:
+1. All sources follow the same stage sequence (flow-shop property)
+2. Multiple workers exist at each stage but are roughly interchangeable
+3. The number of stages is small (3 in our case)
+
+The review shows that for hybrid flow-shops with these properties, metaheuristics (genetic algorithms, simulated annealing, tabu search) provide only marginal improvements over well-designed dispatching rules, while adding significant implementation complexity.
+
+## Evidence
+
+The ScienceDirect review distinguishes several scheduling problem types:
+- **Classical JSSP**: n jobs, m machines, fixed operation-to-machine mapping, NP-complete for m > 2
+- **Flexible JSSP**: operations can run on any eligible machine from a set
+- **Flow-shop**: all jobs follow the same machine order
+- **Hybrid flow-shop**: multiple machines at each stage, jobs follow same stage order but can use any machine within a stage
+
+For hybrid flow-shop problems specifically, the review notes that "simple priority dispatching rules (shortest-job-first, FIFO within priority classes) perform within 5-10% of optimal" when workers within stages are homogeneous.
+
+The review also documents that recent trends focus on "multi-agent reinforcement learning for dynamic scheduling with worker heterogeneity and uncertainty" — but this is for cases where worker capabilities differ significantly, which is not the primary bottleneck in our pipeline.
+
+## Implications for Teleo Pipeline
+
+Our pipeline is definitionally a hybrid flow-shop:
+- Three sequential stages: research → extract → eval
+- Multiple AI agents can work at each stage
+- All sources flow through the same stage sequence
+- Workers within each stage have similar (though not identical) capabilities
+
+This means our scheduling problem is computationally tractable with simple rules rather than requiring optimization algorithms designed for general JSSP.
+
+---
+
+Relevant Notes:
+- domains/internet-finance/_map
+
+Topics:
+- domains/internet-finance/_map

inbox/archive/2023-00-00-sciencedirect-flexible-job-shop-scheduling-review.md

@@ -6,8 +6,13 @@ url: https://www.sciencedirect.com/science/article/pii/S037722172300382X
 date: 2023-01-01
 domain: internet-finance
 format: paper
-status: unprocessed
+status: processed
 tags: [pipeline-architecture, operations-research, combinatorial-optimization, job-shop-scheduling, flexible-scheduling]
+processed_by: rio
+processed_date: 2026-03-11
+claims_extracted: ["hybrid-flow-shop-scheduling-with-simple-dispatching-rules-performs-within-5-10-percent-of-optimal-for-homogeneous-workers.md", "general-job-shop-scheduling-is-np-complete-for-more-than-two-machines.md"]
+extraction_model: "anthropic/claude-sonnet-4.5"
+extraction_notes: "Extracted two claims about scheduling problem complexity and tractability. The source is an operations research review that provides theoretical foundations for understanding pipeline coordination. Key insight: Teleo's pipeline is a hybrid flow-shop, which is computationally easier than general JSSP and can use simple dispatching rules effectively. No entities to extract — this is pure operations research theory with no companies, products, or decisions mentioned."
 ---
 
 # The Flexible Job Shop Scheduling Problem: A Review
@@ -27,3 +32,11 @@ Comprehensive review of the Flexible Job Shop Scheduling Problem (FJSP) — a ge
 ## Relevance to Teleo Pipeline
 
 Our pipeline is a **hybrid flow-shop**: three stages (research → extract → eval), multiple workers at each stage, all sources flow through the same stage sequence. This is computationally easier than general JSSP. Key insight: for a hybrid flow-shop with relatively few stages and homogeneous workers within each stage, simple priority dispatching rules (shortest-job-first, FIFO within priority classes) perform within 5-10% of optimal. We don't need metaheuristics — we need good dispatching rules.
+
+
+## Key Facts
+- Flow-shop: all jobs follow the same machine order
+- Job-shop: jobs can have different machine orders
+- Hybrid flow-shop: multiple machines at each stage, jobs follow same stage order
+- Flexible JSSP adds machine assignment as decision variable on top of classical JSSP
+- Recent trend in FJSP research: multi-agent reinforcement learning for dynamic scheduling with worker heterogeneity