chronos-mesh

Mathematical Formulation & Operations Research Specification

1. Neuro-Symbolic Architectural Separation

Large Language Models excel at parsing unstructured geopolitical, customs, and supplier force-majeure advisories, but fail catastrophically when asked to directly output feasible multi-echelon shipment quantities under coupled capacity, fixed-charge, and SLA constraints.

ChronosMesh enforces a strict Neuro-Symbolic boundary:

  1. Symbolic Compilation (src/chronos_mesh/neuro/): Unstructured text $\mathcal{T}$ is compiled into a formal parameter patch $\Delta = (\alpha_s, \mathcal{B}{\text{lanes}}, \delta{\ell}, \tau_s)$.
  2. Exact Mixed-Integer Optimization (src/chronos_mesh/optimizer/): A Two-Phase Simplex + Best-First Branch-and-Bound MILP solver computes globally optimal primal allocations $(q^, y^, u^*)$.

2. Fixed-Charge Multi-Sourcing MILP Formulation

Let $\mathcal{L}$ be the set of transport corridors and $\mathcal{S}$ the set of upstream facilities. Decision variables are continuous lane flows $q_\ell \ge 0$, binary charter/corridor activations $y_\ell \in {0, 1}$, and unmet demand slack $u \ge 0$:

\[\min_{q, y, u} \sum_{\ell \in \mathcal{L}} \Big( w_c c_\ell(\Delta) + w_t d_\ell(\Delta) + w_g e_\ell \Big) q_\ell + w_c \sum_{\ell \in \mathcal{L}} F_\ell y_\ell + P \cdot u\]

subject to:

  1. Demand Conservation: $\sum_{\ell \in \mathcal{L}} q_\ell + u = D$
  2. Big-$M$ Fixed-Charge Activation: $q_\ell - M_\ell y_\ell \le 0 \quad \forall \ell \in \mathcal{L}$
  3. Shocked Facility Capacity: $\sum_{\ell \in \text{out}(s)} q_\ell \le \alpha_s K_s \quad \forall s \in \mathcal{S}$
  4. Anti-Concentration Diversification: $\sum_{\ell \in \text{out}(s)} q_\ell \le \theta_{\max} D \quad (\theta_{\max} = 0.65)$
  5. Portfolio SLA Lead-Time Ceiling: $\sum_{\ell \in \mathcal{L}} \big(d_\ell(\Delta) - L_{\max}\big) q_\ell \le 0$
  6. Integrality: $y_\ell \in {0, 1}, \; q_\ell \ge 0, \; u \ge 0$

3. Simchi-Levi Risk Exposure Index & Safety Stock