Three-dimensional integrability for the multispecies ASEP: a tetrahedron-equation route to the Zamolodchikov–Faddeev algebra and the Macdonald connection

This task addresses the open question posed in Section 6 of Kuniba–Okado–Scrimshaw, 'A strange five vertex model and multispecies ASEP on a ring' (2024; Algebraic Combinatorics 2025). The stationary states of the multispecies asymmetric simple exclusion process (ASEP) on a ring are governed by a Zamolodchikov–Faddeev (ZF) algebra of operators built as corner transfer matrices of a t-oscillator-weighted 'strange' five-vertex model. For TASEP (t=0) the corresponding algebra follows non-inductively from the tetrahedron equation of three-dimensional integrability, but for generic asymmetry t≠0 every known derivation, including the corner-transfer-matrix one, proceeds by induction on the species rank. The open problem is whether a quantum-oscillator/CTM framework based on a three-dimensional integrability structure can be formulated for ASEP that yields the ZF algebra without rank induction and smoothly interpolates the TASEP and ASEP constructions. The source's second concluding remark separately notes that a comprehensive treatment of the applications of the Macdonald parameter q — connecting the framework to Macdonald-polynomial identities — to multiline queues and t-oscillators requires further investigation.

Background

The asymmetric simple exclusion process (ASEP) is a stochastic lattice gas in which particles hop between neighbouring sites of a one-dimensional lattice with exclusion (at most one particle per site); the right and left hopping rates are unequal, their ratio being measured by an asymmetry parameter $t$ , and the totally asymmetric exclusion process (TASEP) is the limit $t=0$ . On a ring the Markov chain has a unique stationary distribution, and in the multispecies ( $n$ -species) version, where each particle carries a label in $\{0,1,\dots,n\}$ (label 0 = hole) and only adjacent particles with distinct labels may interchange, an explicit description of this distribution is a central problem of integrable probability. Three equivalent descriptions are known: the matrix product solution of Prolhac–Evans–Mallick, whose operators have a recursive tensor structure in the rank $n$ ; the multiline queues (MLQ) of Ferrari–Martin — combinatorial systems of queues in tandem whose weighted sums give stationary probabilities — generalized from TASEP to ASEP by Martin; and the Zamolodchikov–Faddeev (ZF) algebra approach of Cantini–de Gier–Wheeler, in which operators $X_0(z),\dots,X_n(z)$ obey quadratic relations

$X_\alpha(y)\,X_\beta(x)=\sum_{\gamma,\delta}R(y/x)^{\beta,\alpha}_{\gamma,\delta}\,X_\gamma(x)\,X_\delta(y)$

with structure function $R$ the stochastic $R$ -matrix (a sum-to-unity solution of the Yang–Baxter equation — the commutativity relation for transfer matrices of a two-dimensional vertex model — on symmetric-tensor representations of the quantum affine algebra $U_t(\widehat{\mathfrak{sl}}_{n+1})$ ), from which the stationary probabilities follow as traces and partition functions. These probabilities are moreover connected to Macdonald polynomials — the two-parameter symmetric polynomials $P_\lambda(x;q,t)$ of Macdonald theory — since Cantini–de Gier–Wheeler gave a matrix product formula for them and showed that stationary-state normalizations of the multispecies ASEP are Macdonald polynomials at $q=1$ , and Corteel–Mandelshtam–Williams connected MLQ weights to Macdonald polynomials; the asymmetry parameter $t$ is the Macdonald $t$ , so the missing ingredient on this side is the second Macdonald parameter $q$ .

Kuniba–Okado–Scrimshaw (2024, published 2025) reorganized this theory through a 'strange' five-vertex model: a two-dimensional vertex model with $\{0,1\}$ edge states, five allowed vertex configurations, and Boltzmann weights that are operators of a $t$ -deformed quantum oscillator (a one-parameter deformation of the harmonic-oscillator algebra acting on a Fock space). Its weight conservation is 'strange' — the weights $S^{ab}_{ij}$ vanish unless $a+b=j$ , instead of the conventional $a+b=i+j$ — and is designed around the MLQ pairing rule. In this language the matrix product operators become corner transfer matrices (CTMs), i.e. partition functions of a quadrant of the lattice, the stationary probabilities become partition functions of a three-dimensional system with the oscillator acting in the third direction, and the ZF algebra acquires an intrinsic proof through a rank-reducing $RTT=TTR$ relation (an exchange relation that passes the stochastic $R$ -matrix through products of the matrix-product operators, linked to the Yang–Baxter equation) — but, like all earlier ASEP derivations, still by induction on the rank $n$ .

For TASEP the structure is better: the conventional five-vertex model of Kuniba–Maruyama–Okado embeds the problem into a genuine three-dimensional integrable system whose local $L$ -operators (of Bazhanov–Sergeev type, built on $q$ -oscillator representations, where this $q$ is the oscillator deformation parameter — the rôle played by the asymmetry $t$ — and is not the Macdonald parameter $q$ ) satisfy the tetrahedron equation — the three-dimensional generalization of the Yang–Baxter equation that guarantees commutativity of layer-to-layer transfer matrices — and this equation reduces the ZF algebra directly, with no induction on $n$ . The two five-vertex models are related only by an index shift that trades one weight-conservation law for the other, and the source paper records that although the TASEP and ASEP ZF algebras are smoothly connected via $t$ , the authors found no method to formulate the ASEP results on the conventional model. Whether a quantum-oscillator/CTM framework for ASEP exists that is genuinely three-dimensional integrable — so that a tetrahedron-type equation, not rank induction, produces the ZF algebra, with smooth interpolation between $t=0$ and $t\neq0$ — is the open question posed here.

Problem Statement

The problem, posed in Section 6 of Kuniba–Okado–Scrimshaw, is to complete the quantum-oscillator/CTM framework for the stationary states of the $n$ -species ASEP on a ring into a genuinely three-dimensional integrable structure; this is the open question of the source's first concluding remark. The source's second concluding remark separately notes a related open direction along the Macdonald parameter $q$ ; it is recorded below for context but is not part of the completion criterion for the open question.

(1) Core question ( $t$ -direction). Determine whether there exists a variant of the quantum-oscillator/CTM approach for the multispecies ASEP at generic asymmetry $t\neq0$ , expected to realize a three-dimensional integrability structure (a family of $L$ -operators or layer-to-layer transfer matrices satisfying a tetrahedron-type equation), such that: (a) the stationary probabilities of the $n$ -species ASEP are reproduced as partition functions or trace formulas of the construction, agreeing with the matrix product of Prolhac–Evans–Mallick and the multiline-queue weights of Martin and of Corteel–Mandelshtam–Williams; (b) the Zamolodchikov–Faddeev algebra relations among the CTM operators are derived from the tetrahedron-type relation directly, without induction on the rank $n$ ; (c) at $t=0$ the construction reduces to the tetrahedron-equation construction of Kuniba–Maruyama–Okado for TASEP, so that the two constructions are smoothly interpolated by the parameter $t$ ; formulating the ASEP results within the conventional weight-conserving five-vertex model on which the tetrahedron equation acts is one expected route to this interpolation, not an additional requirement.

(2) Related open direction ( $q$ -direction), noted in the source's second concluding remark but not part of the completion criterion for (1): incorporate, within the same framework or an explicitly extended one, the supplementary Macdonald parameter $q$ and the weight variables so that the known ASEP–Macdonald connections — the matrix product formula for Macdonald polynomials and the identification of stationary-state normalizations with Macdonald polynomials at $q=1$ — are derived systematically; the source states that a comprehensive treatment of the applications of $q$ to MLQs and $t$ -oscillators 'requires further investigation'.

Component (1) is the open question posed by the source ('It remains an open question whether such a superior variant of the quantum oscillator/CTM approach can be formulated that smoothly interpolates TASEP and ASEP'); a complete answer resolves (1) — either by exhibiting a construction satisfying (a)–(c) or by proving that no such construction exists. Component (2) is an independent future-work note of the source, with its own softer scope, and is not a mandatory joint deliverable. Regarding (1), the source also flags Martin's alternative queueing construction (Section 7 of Martin 2020) as a possible guide ('may offer valuable insights').

References

Atsuo Kuniba, Masato Okado, Travis Scrimshaw, A strange five vertex model and multispecies ASEP on a ring, Algebraic Combinatorics 8(6) (2025) 1713–1741, DOI: 10.5802/alco.457; arXiv:2408.12092, https://arxiv.org/abs/2408.12092.
Atsuo Kuniba, Shouya Maruyama, Masato Okado, Multispecies TASEP and the tetrahedron equation, Journal of Physics A: Mathematical and Theoretical 49 (2016) 114001, DOI: 10.1088/1751-8113/49/11/114001; arXiv:1509.09018, https://arxiv.org/abs/1509.09018
Atsuo Kuniba, Shouya Maruyama, Masato Okado, Multispecies TASEP and combinatorial R, Journal of Physics A: Mathematical and Theoretical 48 (2015) 34FT02, DOI: 10.1088/1751-8113/48/34/34FT02; arXiv:1506.04490, https://arxiv.org/abs/1506.04490
Atsuo Kuniba, Quantum Groups in Three-Dimensional Integrability, Springer, Singapore, 2022, DOI: 10.1007/978-981-19-3262-5, ISBN 978-981-19-3261-8, https://doi.org/10.1007/978-981-19-3262-5
V. V. Bazhanov, S. M. Sergeev, Zamolodchikov's tetrahedron equation and hidden structure of quantum groups, Journal of Physics A: Mathematical and General 39 (2006) 3295–3310, DOI: 10.1088/0305-4470/39/13/009, https://doi.org/10.1088/0305-4470/39/13/009
A. Kuniba, V. V. Mangazeev, S. Maruyama, M. Okado, Stochastic R matrix for $U_q(A_n^{(1)})$ , Nuclear Physics B 913 (2016) 248–277, DOI: 10.1016/j.nuclphysb.2016.09.016; arXiv:1604.08304, https://arxiv.org/abs/1604.08304
S. Prolhac, M. R. Evans, K. Mallick, The matrix product solution of the multispecies partially asymmetric exclusion process, Journal of Physics A: Mathematical and Theoretical 42 (2009) 165004, DOI: 10.1088/1751-8113/42/16/165004; arXiv:0812.3293, https://arxiv.org/abs/0812.3293
Pablo A. Ferrari, James B. Martin, Stationary distributions of multi-type totally asymmetric exclusion processes, The Annals of Probability 35(3) (2007) 807–832, DOI: 10.1214/009117906000000944; arXiv:math/0501291, https://arxiv.org/abs/math/0501291
James B. Martin, Stationary distributions of the multi-type ASEP, Electronic Journal of Probability 25 (2020) Paper No. 43, DOI: 10.1214/20-ejp421; arXiv:1810.10650, https://arxiv.org/abs/1810.10650
Luigi Cantini, Jan de Gier, Michael Wheeler, Matrix product formula for Macdonald polynomials, Journal of Physics A: Mathematical and Theoretical 48 (2015) 384001, DOI: 10.1088/1751-8113/48/38/384001; arXiv:1505.00287, https://arxiv.org/abs/1505.00287
Sylvie Corteel, Olya Mandelshtam, Lauren Williams, From multiline queues to Macdonald polynomials via the exclusion process, American Journal of Mathematics 144(2) (2022) 395–436, DOI: 10.1353/ajm.2022.0007; arXiv:1811.01024, https://arxiv.org/abs/1811.01024
Amol Aggarwal, Matthew Nicoletti, Leonid Petrov, Colored interacting particle systems on the ring: Stationary measures from Yang–Baxter equations, Compositio Mathematica 161 (2025) 1855–1922, DOI: 10.1112/s0010437x25007195; arXiv:2309.11865, https://arxiv.org/abs/2309.11865
Arvind Ayyer, Atsuo Kuniba, Multispecies inhomogeneous $t$ -PushTASEP from antisymmetric fusion, Electronic Journal of Probability 30 (2025), DOI: 10.1214/25-EJP1448; arXiv:2503.00829, https://arxiv.org/abs/2503.00829
Arvind Ayyer, Atsuo Kuniba, Multispecies inhomogeneous $t$ -PushTASEP with general capacity, arXiv:2602.17179 (2026), https://arxiv.org/abs/2602.17179
Shinsuke Iwao, Kohei Motegi, Ryo Ohkawa, Tetrahedron equation and Schur functions, Journal of Physics A: Mathematical and Theoretical 58 (2025) 015201, DOI: 10.1088/1751-8121/ad9675; arXiv:2405.10011, https://arxiv.org/abs/2405.10011
Shinsuke Iwao, Kohei Motegi, Ryo Ohkawa, Tetrahedral $L$ -operators, tensor Schur polynomials and $q$ -deformed loop elementary symmetric functions, arXiv:2604.22141 (2026), https://arxiv.org/abs/2604.22141
Olya Mandelshtam, Jerónimo Valencia-Porras, Type $C$ multiline queues and the open-boundary TASEP, arXiv:2607.24637 (2026), https://arxiv.org/abs/2607.24637
Arvind Ayyer, James B. Martin, Lauren Williams, The inhomogeneous $t$ -PushTASEP and Macdonald polynomials at $q=1$ , Annales de l'Institut Henri Poincaré D, Combinatorics, Physics and their Interactions (2025), DOI: 10.4171/aihpd/210; arXiv:2403.10485, https://arxiv.org/abs/2403.10485
Olya Mandelshtam, Jerónimo Valencia-Porras, Macdonald polynomials at $t=0$ through twisted multiline queues, Advances in Applied Mathematics 174 (2026) 103020, DOI: 10.1016/j.aam.2025.103020; arXiv:2407.05362, https://arxiv.org/abs/2407.05362
Olya Mandelshtam, New formulas for Macdonald polynomials via the multispecies exclusion and zero range processes, in: Macdonald Theory and Beyond, Contemporary Mathematics 815, American Mathematical Society (2025) 91–114, DOI: 10.1090/conm/815/16316; arXiv:2508.04666, https://arxiv.org/abs/2508.04666