{"id":"2677a3f2-fa4a-4cf4-83a1-c2258c3fcce2","arxiv_id":"1908.05995","paper_version":1,"verdict":"ACCEPT","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"The paper derives input-to-state stability type estimates for the 1-D continuity equation in every Lp norm with p>1, including the sup norm, and applies them to a nonlinear manufacturing model.","lead":"This paper proves stability estimates for the 1-D continuity equation when the velocity and boundary influx are treated as external inputs. The estimates hold in all Lp norms and are then applied to feedback control of manufacturing models.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Theorem 2.1 invokes Theorem 2.3 with a = -v_x, but Theorem 2.3 assumes a in C^1 while Theorem 2.1 only assumes v in C^1; the proof as written has a regularity gap.","rationale":"The reader's weakest assumption is positivity of v, which is indeed needed for the characteristic structure and the finite look-back window. My concern is different and more directly about the proof: Theorem 2.1 says v in C^1, but the auxiliary theorem it invokes requires the coefficient a = -v_x to be C^1. This is a real mismatch in the supplied text. I do not think the main estimates are false: the characteristic and Lyapunov arguments for (2.10) and (2.11) appear to use only boundedness of a, and the special structure a = -v_x preserves C^1 regularity of rho. However, the paper does not say this, so the proof as written is incomplete at a load-bearing point. The requested check is a focused verification of whether Theorem 2.3's estimates require a in C^1 or only a in C^0. If the weaker hypothesis suffices, the correct remedy is a small revision to the statements, not a change of the mathematical conclusions. Hence I recommend conditional acceptance rather than rejection or unqualified acceptance.","tokens_in":27784,"tokens_out":36844,"duration_ms":368022,"concrete_test":"Re-derive Theorem 2.3 with the hypothesis a in C^0(R_+ x [0,1]) (keeping the other assumptions). If (2.10) and (2.11) still follow from the existing proofs, then the patch is purely to weaken the statement and Theorem 2.1 is sound. If any step in the proof genuinely needs differentiability of a, identify that step and either add v in C^2 to Theorem 2.1 or supply the missing argument for a = -v_x.","verdict_should_be":"CONDITIONAL","load_bearing_attack":"The central proof of Theorem 2.1 is described as an application of Theorem 2.3 to w = ln(rho/rho_s). For that transformation, the linear PDE for w has a(t,x) = -v_x(t,x). Theorem 2.3 is stated for a in C^1(R_+ x [0,1]), but Theorem 2.1 assumes only v in C^1([0,1] x R_+; (0, infinity)), so -v_x is merely continuous. Thus the hypotheses of Theorem 2.3 are not met as typeset. This is not a counterexample to the estimates: the derivation of (2.10) and (2.11) through characteristic formulas and the Lyapunov estimate only uses A(t) = max |a| over the domain and never differentiates a, so the estimates should extend to continuous a. But that extension is not stated or proved, and the C^1 regularity conclusion for rho also needs a separate argument from the conservative form when v is only C^1. As written, a reader cannot formally close the main theorem from the stated auxiliary theorem.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper studies ISS-type stability estimates for the one-dimensional continuity equation ρ_t + (vρ)_x = 0 on [0,1], with positive time-varying velocity v as a distributed input and an exponential boundary disturbance b. The main result, Theorem 2.1, bounds the Lp norm (p>1) and the sup norm of the logarithmic deviation ln(ρ/ρ_s) by the initial deviation plus explicit terms involving the maximum of |v_x| and the boundary disturbance over a finite look-back interval. The proof is based on a linear transport lemma, Theorem 2.3, proved by characteristics, superposition, and a Lyapunov functional. Section 3 applies these estimates to a nonlocal manufacturing model under feedback control, yielding Theorem 3.1 via a contraction argument for the nonlocal velocity. The paper also discusses sharpness and the dependence of gains on the velocity.","tokens_in":28020,"tokens_out":10566,"duration_ms":106122,"significance":"If the stated estimates are valid, the paper provides explicit finite-time-memory stability estimates for a bilinear transport equation with time-varying velocity, which are not standard ISS estimates but are potentially useful in small-gain analysis of larger physical models. The application to nonlocal manufacturing models is nontrivial and the fixed-point proof in Proposition 4.1 is a genuine contribution. The paper is honest about the non-ISS features and includes a valuable discussion of sharpness in Remark 2.2. However, two technical points in the proof of the main results need to be fixed before the claims can be accepted as stated.","major_comments":[{"comment":"Theorem 2.1 assumes v ∈ C^1 and applies Theorem 2.3 with a = -v_x and f = 0. However, Theorem 2.3 is stated for a ∈ C^1, whereas -v_x is only continuous under the hypothesis v ∈ C^1. This is not a purely cosmetic mismatch: the regularity conclusion of Theorem 2.3 (that w, and hence ρ, is C^1) relies on differentiating the characteristic formula, which requires some differentiability of a in x. With only v_x continuous, the solution ρ of the conservative continuity equation need not be C^1; for example, v(x)=x^{4/3} on [0,1] is C^1 but produces characteristics whose solution has non-continuous first derivative at x=0. Thus the stated theorem cannot be formally closed from Theorem 2.3 as written. The estimates may still hold for a suitable weaker solution class or under v ∈ C^2, but the manuscript must either strengthen the hypothesis of Theorem 2.1, weaken the regularity claim, or extend Theorem 2.3 to continuous coefficients with a complete proof of the required regularity and estimates.","section":"Proof of Theorem 2.1 / Theorem 2.3"},{"comment":"The sup-norm estimate (2.11) is claimed to follow by letting p → ∞ in (2.10). This is not justified as written: the constants in (2.10), and in particular the factor exp((p v_max(t)+A(t))/v_min(t)) appearing in the estimates derived from (4.25) and (4.13), grow with p and do not converge to finite limits. Consequently, taking p → ∞ does not produce a finite, p-independent right-hand side. A separate argument for the sup norm, e.g., a direct characteristic estimate for ||w(t)||_∞, is needed. This issue affects the sup-norm claims in Theorems 2.1 and 3.1 as well.","section":"Proof of Theorem 2.3, derivation of (2.11)"}],"minor_comments":[{"comment":"Equations (2.4), (2.5), (2.10), (2.11), (3.6) and (3.7) are extremely difficult to read because of compressed notation and missing grouping symbols; please reformat them so that exponents, maxima, and arguments of h are unambiguous.","section":"Throughout, displayed estimates"},{"comment":"The function spaces for a, f, and v are stated with garbled notation; please state explicitly, for example, a ∈ C^1(R_+ × [0,1]), f ∈ C^1(R_+ × [0,1]), and v ∈ C^1(R_+ × [0,1]; (0,∞)).","section":"Theorem 2.3 statement"},{"comment":"The compatibility conditions at (0,0) are written in compressed form; please expand them fully (e.g., ρ_0(0) = ρ_s exp(b(0)) and the corresponding first-order compatibility condition) to make the dependency on v and b explicit.","section":"Theorem 2.1 and Theorem 3.1, compatibility conditions"},{"comment":"The function h is defined as h(s)=1 for s<0 and h(s)=0 for s≥0, but in several displayed estimates its argument is not legible; make the argument explicit, for example h(t - 1/v_min(t)).","section":"Definition of h"}],"recommendation":"major_revision","confidential_remarks":"The regularity gap and the p→∞ passage are substantive but likely fixable within the scope of the paper. If the authors can either require v∈C^2 for the C^1 solution statement or extend Theorem 2.3 to continuous a with a separate sup-norm proof, the paper would be acceptable. The core estimates and the manufacturing application are interesting and worth publishing after revision."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"This paper does something genuinely useful: it gives explicit stability estimates for the 1-D continuity equation in every Lp norm and the sup norm, with boundary disturbances, and it does so via characteristics and a Lyapunov functional. The estimates have finite-time memory and velocity-dependent gains, which is the right kind of statement for a bilinear transport PDE. The manufacturing application is a legitimate bonus and shows the estimates are not a dead end. I agree with the reader that the core results are new and correctly derived. The work deserves a serious referee.\n\nThe soft spot that matters is real but small. Theorem 2.1 states v is only C^1, but the proof applies Theorem 2.3 with a = -v_x. Theorem 2.3 requires a to be C^1, and v_x is merely continuous under the stated assumption. So as typeset, the proof of Theorem 2.1 does not formally close. The stress-test note has it right. That said, I checked the proof of Theorem 2.3: the estimates only use A(t) = max |a| over a rectangle and never differentiate a. The same argument works for continuous a. So the gap is a hypothesis mismatch, not a substantive error. The authors could either weaken Theorem 2.3 to a ∈ C^0 or strengthen Theorem 2.1 to v ∈ C^2. Neither change affects the statements' meaning. This is exactly the kind of thing a referee should insist be cleaned up, not a reason to reject.\n\nOther limitations are stated honestly and are not red flags. The positivity of the state and strict positivity of v are used heavily and are acknowledged. The claim that these are ISS-like but not exactly ISS because gains depend on the input is handled openly. The self-citation to Lemma 2.12 in the authors' own book is fine; it is a standard comparison lemma and the dependency is visible. I could not fully verify every OCR-corrupted inequality line, but the structure of the characteristic formulas and the Lyapunov derivation is clear enough that I found no hidden circularity or invented entities.\n\nThe right readership is people working on control of transport PDEs, age-structured models, and nonlocal conservation laws. For that audience this is a reference-grade set of estimates. I would send this to peer review with a request to fix the regularity mismatch in the proof of Theorem 2.1, and I would cite it in my own work.","headline":"Solid, useful ISS-type estimates for the 1-D continuity equation; the main theorem has a small but real regularity gap in its proof that is easy to fix.","tokens_in":28508,"tokens_out":1269,"would_cite":true,"duration_ms":14685,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":["35L60","93D25","93D05"],"pacs":[],"model":"deepseek-v4-flash","headline":"The 1-D continuity equation admits explicit stability bounds in every Lp norm, with p>1 and the sup norm.","keywords":["continuity equation","transport PDE","input-to-state stability","boundary disturbance","logarithmic norm","finite-time stability","Lp norms","manufacturing systems"],"falsifier":"Take a positive $C^1$ velocity and boundary disturbance, such as $v(t,x)=2+\\sin(t)\\,x$ with $b(t)=0.1\\,\\sin(t)$, and compute the exact solution of (2.1)-(2.2) along the characteristic curves by quadrature. Then evaluate both sides of (2.4) for $p=2$ and several values of $\\mu$; a single time at which the inequality fails would disprove Theorem 2.1, while a sweep over many randomly chosen positive velocities and disturbances would provide numerical confirmation that the bounds hold.","tokens_in":27610,"feed_emoji":"⏳","tokens_out":7071,"duration_ms":71669,"temperature":0.7,"pith_summary":"This paper claims that the stability of the 1-D continuity equation, with velocity viewed as an input and with a boundary disturbance, can be quantified by explicit estimates in all $L^p$ norms for $p>1$ and in the sup norm. The state appears through its logarithmic deviation from a constant equilibrium, and the bounds involve only the initial deviation plus the largest values of the velocity gradient and boundary disturbance over a finite, velocity-dependent time window in the past. This matters because the continuity equation is the conservation-law core of fluid, traffic, and manufacturing models, and the estimates give explicit input-to-state-type stability bounds for the bilinear equation with time-varying velocity and boundary disturbance. The paper also shows how the estimates apply directly to the feedback stabilization of a non-local manufacturing model.","feed_headline":"A single travel-time window governs 1-D continuity stability","feed_subtitle":"The paper gives explicit Lp and sup-norm estimates in which velocity and boundary disturbances enter through a finite look-back interval.","key_machinery":"The argument runs through the change of variable $w=\\ln(\\rho/\\rho_s)$, which turns the bilinear continuity equation into the linear transport equation (2.7) with coefficient $a=-v_x$. The solution is then split into three components by superposition: the initial profile carried along characteristics, the boundary input $b$, and the distributed source. The first two components are bounded using exact characteristic formulas; the third component is bounded with an exponentially weighted Lyapunov functional whose differential inequality is closed by a comparison lemma. This yields the finite-memory estimates (2.10) and (2.11), and Theorem 2.1 follows by substituting $a=-v_x$ and $f=0$. For the manufacturing model, the proof uses a contraction-mapping fixed-point argument to supply the a priori bounds needed before Theorem 2.1 is applied.","core_discovery":"The central claim is Theorem 2.1: for a positive $C^1$ velocity $v(t,x)$ and a boundary disturbance $b(t)$ at the inflow boundary $x=0$, the logarithmic deviation $w=\\ln(\\rho/\\rho_s)$ obeys the quantitative estimates (2.4) and (2.5). For any $p>1$ and a suitable positive parameter $\\mu$, the $L^p$ norm and the sup norm of $w(t,\\cdot)$ are bounded by the corresponding norm of the initial deviation plus a term involving the sup of $|v_x|$ over the look-back interval and a term involving the sup of $|b|$ over the same interval, both weighted by $\\exp(-\\mu(t-s))$. The coefficients, or gains, depend on the velocity and its minimum, so the estimates are not classical ISS estimates; they express a finite-memory, relative-error type of stability in the logarithmic norm. In the undisturbed constant-velocity case, both estimates give finite-time stability. Theorem 3.1 transfers these bounds to a non-local nonlinear manufacturing model under the feedback law $u(t)=\\lambda(W(t))\\,\\rho_s\\exp(b(t))$, yielding estimates (3.6) and (3.7).","pith_inferences":["The characteristic-plus-Lyapunov decomposition in Theorem 2.3 should generalize to systems of one-dimensional balance laws with several transport speeds, since the boundary and source estimates are per-characteristic; deriving the analogue of (2.10) for such systems would be a natural test of the method.","The finite-memory window suggests that in interconnected networks the continuity equation can be treated, for stability purposes, as a delay system whose delay is the instantaneous traversal time $1/v_{\\min}(t)$, which could simplify feedback designs for transportation and production networks.","Because the estimates control the logarithmic norm, they measure relative error of the density; a positivity-preserving numerical scheme could use these bounds to certify that a computed solution stays within a prescribed relative accuracy.","The same method may apply to non-local traffic models, where the velocity depends on the density integrated over space, since the manufacturing application already demonstrates how to handle a spatially constant but time-varying velocity of that form."],"forward_implications":["If the estimates are correct, a conservation-law subsystem of a larger model can be treated as a block with known input-to-state behavior, opening the way to small-gain stability proofs for networks of transport and conservation laws.","For constant velocity and no disturbance, the estimates imply that the logarithmic deviation reaches zero after the travel time across the domain, recovering finite-time stability in the log norm.","Only the velocity and boundary history over the interval $[\\max(0,t-1/v_{\\min}(t)),\\,t]$ can influence the state at time $t$; any older input has no effect on the bound.","Velocities that increase with $x$ produce a strictly larger bias in the density profile than velocities that decrease with $x$, as quantified by the velocity-gradient term in the estimates.","For the non-local manufacturing model, the feedback law is robust to bounded boundary uncertainty $b$, and finite-time stability holds when $b=0$, although the settling time can become large for large disturbances or large initial densities."],"supporting_citations":[{"why":"Supplies the comparison lemma used to turn the Lyapunov differential inequality for the distributed-source component into the finite-memory integral estimate in Theorem 2.3.","marker":"[16]"},{"why":"Introduces the highly re-entrant manufacturing conservation-law model that Section 3 treats, providing the starting point for the feedback-stabilization application.","marker":"[7]"},{"why":"Provides the controllability analysis for a scalar conservation law with nonlocal velocity, one of the model foundations for the manufacturing system studied in Theorem 3.1.","marker":"[8]"},{"why":"Gives output feedback stabilization results for a scalar conservation law with nonlocal velocity, supplying context for the feedback law (3.4).","marker":"[9]"},{"why":"Presents continuum models of production systems whose nonlocal structure is the basis for the manufacturing model analyzed here.","marker":"[23]"}],"fun_headline_variants":["Velocity-dependent ISS gains bound continuity flow","Look-back interval controls 1-D continuity stability","Log-norm estimates set continuity stability window","Finite-memory proof stabilizes nonlinear manufacturing","Travel-time window governs continuity equation stability"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The argument collapses if the velocity is allowed to vanish or change sign: Theorem 2.1 assumes $v(t,x)$ is $C^1$ and strictly positive on $[0,1]$ for all times, which makes $x=0$ the only inflow boundary and keeps the finite-memory interval $[\\max(0,t-1/v_{\\min}(t)),\\,t]$ well defined.","fun_headline_variants_meta":{"raw":{"variants":["Velocity-dependent ISS gains bound continuity flow","Look-back interval controls 1-D continuity stability","Log-norm estimates set continuity stability window","Finite-memory proof stabilizes nonlinear manufacturing","Travel-time window governs continuity equation stability"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000157,"raw_usage":{"total_tokens":1214,"prompt_tokens":932,"completion_tokens":282,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":548,"completion_tokens_details":{"reasoning_tokens":216}},"tokens_in":548,"tokens_out":282,"duration_ms":3538,"temperature":1.0,"reasoning_tokens":216,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-14T12:57:43.125849+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Take a positive $C^1$ velocity and boundary disturbance, such as $v(t,x)=2+\\sin(t)\\,x$ with $b(t)=0.1\\,\\sin(t)$, and compute the exact solution of (2.1)-(2.2) along the characteristic curves by quadrature. Then evaluate both sides of (2.4) for $p=2$ and several values of $\\mu$; a single time at which the inequality fails would disprove Theorem 2.1, while a sweep over many randomly chosen positive velocities and disturbances would provide numerical confirmation that the bounds hold.","supporting_citations":[{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Supplies the comparison lemma used to turn the Lyapunov differential inequality for the distributed-source component into the finite-memory integral estimate in Theorem 2.3."},{"cited_title":"Analysis of a Conservation Law Modeling a Highly Re- entrant Manufacturing System","cited_arxiv_id":null,"evidence_quote":"Introduces the highly re-entrant manufacturing conservation-law model that Section 3 treats, providing the starting point for the feedback-stabilization application."},{"cited_title":"Controllability for a S calar Conservation La w with Nonlocal Velocity","cited_arxiv_id":null,"evidence_quote":"Provides the controllability analysis for a scalar conservation law with nonlocal velocity, one of the model foundations for the manufacturing system studied in Theorem 3.1."},{"cited_title":"Output Feedback Stabilization for a Scalar Conservation Law with a Nonlocal Velocity","cited_arxiv_id":null,"evidence_quote":"Gives output feedback stabilization results for a scalar conservation law with nonlocal velocity, supplying context for the feedback law (3.4)."},{"cited_title":"Control of Continuum Models of Production Systems","cited_arxiv_id":null,"evidence_quote":"Presents continuum models of production systems whose nonlocal structure is the basis for the manufacturing model analyzed here."}],"review_version":1}