{"id":"2611231e-2abd-47fc-bd1f-cd9c4bfafcf3","arxiv_id":"2509.02246","paper_version":1,"verdict":"CONDITIONAL","confidence":"HIGH","novelty_score":4.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":9,"one_line_summary":"A generic rolling access model for synchrotron beamtime, tested at five PETRA III beamlines, could cut submission-to-beamtime wait times from months to weeks.","lead":"The paper describes a rolling, deadline-free system for submitting and reviewing synchrotron beamtime proposals, replacing periodic calls. A synthetic scheduling simulation suggests waiting times could drop to about two months, but the system is still in a test phase at DESY's PETRA III.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Simulation evidence is internally inconsistent: Section 1.9 states 450 virtual proposals, but the Figure 3 caption says 110, and this directly affects the reported wait-time results.","rationale":"The reader correctly identified the simulation's synthetic inputs and the 450 vs 110 inconsistency as weaknesses, and issued a CONDITIONAL verdict. My stress-test focuses on the internal inconsistency as the single most load-bearing concern because it is concrete, directly tied to the paper's central quantitative claim, and currently makes the simulation irreproducible. This does not change the overall verdict: the paper still proposes a plausible operational model with a real test phase underway, so conditional acceptance remains appropriate until the simulation is clarified and validated. My concrete test would settle whether the inconsistency matters for the headline numbers. I partially agree with the reader's weakest_assumption: the reader emphasized the use of synthetic input distributions, whereas I emphasize the internal contradiction in the number of simulated proposals, which is a more immediate and definitive flaw. Both concerns point in the same direction: the quantitative support for 'significantly reduced waiting times' is not yet solid.","tokens_in":8270,"tokens_out":3526,"duration_ms":42916,"concrete_test":"Re-run the scheduling simulation using the exact algorithm described in Section 1.9 with both N=450 and N=110 proposals over 730 days, for both case-1 and case-2 parameter sets. Compare the average waiting time, the waiting time of top-ranked proposals, and the unscheduled-days fraction. If the results differ materially between N=450 and N=110, the reported numbers are not robust and the central claim is unsupported; if the results are essentially identical, the 450 vs 110 discrepancy is a harmless typo and the concern would be resolved. Ideally, the authors should release the simulation code or a complete pseudocode specification so this check can be performed independently.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central quantitative claim—that rolling access reduces waiting times to about two months on average, with top-ranked proposals scheduled in 15–40 days—rests entirely on the Section 1.9 simulation. That simulation is irreproducible as reported: the text says 'A total of 450 virtual beamtime requests were randomly distributed over 730 days,' while the Figure 3 caption states 'The simulation has been performed with 110 project proposals.' These are not interchangeable. With 450 proposals and the stated 3–18 shift range, the total requested load is roughly 4725 shifts; with 110 proposals it is about 1155 shifts. The scheduling horizon in case 1 (30/60/90 days with fill factors 1/0.3/0.1) provides roughly 57 scheduled days per 180-day cycle, i.e. about 171 shifts, or ~694 shifts over 730 days. Thus 450 proposals heavily oversubscribe the modeled capacity, while 110 proposals also oversubscribe but far less severely. The waiting-time distribution, the fraction of proposals cut by one day, and the 2%/3.3% unscheduled-day figures would all be sensitive to this load difference. The paper provides no code, no full algorithmic specification, and no error bars, so the reader cannot determine which scenario produced Figures 3c/3d. Additionally, the simulation excludes review time and removes unscheduled proposals after c=3 months, which caps scheduled waiting times at about three months; the 'average two months' is therefore partially an artifact of that removal rule, not an independent empirical validation. The headline claim about the conventional model's nine-month wait is not reproduced in the same simulation, so no controlled comparison is made. These issues together make the simulation evidence for the paper's primary claim unreliable until the inconsistency is resolved.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper proposes a generic rolling access model for synchrotron radiation beamtime, intended to replace the conventional call-based access model. In the proposed scheme, proposal submission, external review, and scheduling all operate continuously, with a dynamic ranking stack, a three-period filling-factor schedule, and a validity/removal rule. The authors claim that this approach significantly reduces waiting times between proposal submission and experiment execution, and they support this with a simulation of two parameterized scheduling cases (Section 1.9). The manuscript also describes a test phase at five PETRA III beamlines and discusses advantages (unified submission, flexibility, multiple beamtimes) and challenges (dynamic ranking, near-cutoff proposals).","tokens_in":8681,"tokens_out":3444,"duration_ms":40111,"significance":"If the quantitative claims were robust, the model could have practical impact on user access at synchrotron facilities, addressing well-documented delays in the conventional call-based model. The paper's conceptual contribution—a unified rolling framework with explicit scheduling and review procedures—is useful and timely, and the planned test phase at PETRA III is a strength. The authors also make the model's parameters explicit, which is a good starting point for simulation and comparison. However, the current simulation evidence is not sufficient to establish the headline reduction in waiting times; the reported numbers contain a direct internal inconsistency and the results are largely predetermined by the selected scheduling and removal parameters. The central claim is therefore plausible but not yet demonstrated at the level of rigor expected for a quantitative performance claim.","major_comments":[{"comment":"The simulation evidence is internally inconsistent: Section 1.9 states 'A total of 450 virtual beamtime requests were randomly distributed over 730 days,' whereas the Figure 3 caption says 'The simulation has been performed with 110 project proposals.' With 3–18 shifts per request, these imply total requested loads of roughly 4,700 and 1,150 shifts, respectively, against a horizon that provides on the order of 700 shifts over 730 days under the stated filling factors (case 1: 30 days at f=1, 60 days at f=0.3, 90 days at f=0.1 repeated). The waiting-time distributions, the fraction of sessions cut by one day, and the 2%/3.3% unscheduled-days figures are all sensitive to this load difference. The authors must correct this contradiction and provide the exact algorithmic specification or code; otherwise the reported quantitative claims cannot be reproduced or checked.","section":"§1.9, Figure 3 caption"},{"comment":"The reported result that top-ranked proposals are scheduled within 15–40 days is forced by construction: the first scheduling period Δt1 has filling factor f1=1, so the highest-ranked proposals in the stack are necessarily scheduled into that first period. Similarly, the removal rule c=3 months caps the scheduled waiting time at about three months, making the 'average wait time of two months' largely an artifact of the chosen parameters rather than an independent finding. The paper should report waiting times as a function of c and f1, and clearly separate 'time to scheduling' from 'total time from submission to execution,' since the simulation explicitly excludes review time.","section":"§1.9, scheduling rules"},{"comment":"The simulation uses synthetic input distributions—uniform proposal arrivals over 730 days, normally distributed ratings with mean 2.5 and standard deviation 1.0, requested shifts between 3 and 18—with no error bars, no sensitivity analysis, and no comparison with real queue data from the five PETRA III test beamlines. The central quantitative claim is therefore not yet an empirical prediction; it is an illustration conditioned on parameters that have not been validated. A sensitivity analysis over plausible arrival rates, rating distributions, and request sizes is required before the wait-time reduction can be claimed as a general property of the rolling access model.","section":"§1.9, simulation inputs and sensitivity"},{"comment":"The two simulation cases (case 1 and case 2) are compared only through two selected parameter sets, with no statistical characterization of the outputs. Statements such as 'the higher-ranked proposals could be scheduled within less than 40 days' and '3.3% of all days remain unscheduled' are point estimates from a single stochastic run; without multiple runs or confidence intervals, these differences (2% vs 3.3%) cannot be interpreted as reliable findings. The paper should provide summary statistics over multiple seeds or a deterministic specification if the scheduling process is fully deterministic.","section":"§1.9, Figure 3c/3d"}],"minor_comments":[{"comment":"Grammar: 'This significantly reduces the waiting times ... than that of the call-based access model' should be 'compared with' or 'relative to.'","section":"Abstract and §1.2"},{"comment":"The text says 'Figures 3b and 3c illustrate the results for the two case scenarios with correlation plots,' but the correlation plots are actually (c) and (d), while (b) is a schedule example. Please renumber or correct the cross-references.","section":"§1.9, text vs. figures"},{"comment":"Typo: 'Commercial customers purchase beamtime through a contract and are receive priority scheduling' should be 'and receive priority scheduling.'","section":"Second page, first paragraph"},{"comment":"The caption does not state the assumed arrival process, rating distribution, or removal rule, even though these are essential to interpreting the plots. Please add these details.","section":"Figure 3 caption"},{"comment":"SPring-8 is commonly written 'SPring-8' (not 'Spring-8'); please check and standardize facility names throughout.","section":"Page 3, facility name"}],"recommendation":"major_revision","confidential_remarks":"The core idea is potentially useful for the accelerator/facility operations community, but the simulation inconsistency and the parameter-driven nature of the quantitative claims are significant. In my view, the authors should be asked to provide a corrected, reproducible simulation (ideally with code or a complete pseudo-code specification), to add sensitivity analyses, and to reframe the wait-time numbers as illustrative until validated by the PETRA III test phase. I would not reject the manuscript, because the conceptual model is worth publishing if properly qualified."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"The paper is a practical proposal from DESY to replace call-based beamtime access with a rolling model across all beamline types. The framework is sensible and clearly described, and the test phase at five PETRA III beamlines is the right way to validate it. But the central claim—that waiting times drop to about two months—rests on a synthetic simulation that is internally inconsistent (450 proposals in the text, 110 in the Figure 3 caption) and parameter-driven. The quantitative headline should not be taken at face value.\n\nWhat is actually new is generalizing rolling access beyond the automated, standardized beamlines at ESRF and the frequent-call models at Spring-8 and Elettra. The paper's unified submission scheme, dynamic ranking procedure, and three-period scheduling heuristic with filling factors are a legitimate extension. The scheduling rule is simple and reproducible in principle, and the authors are honest that this is a first step with real test-phase data still to come. That gives the work genuine value as an operations proposal.\n\nThe simulation evidence is the soft spot. The 450-versus-110 discrepancy changes the simulated load by roughly a factor of four, and it is not dismissed as a typo. No code, no error bars, no real queue data, and the authors explicitly exclude review time from the wait estimates. The fast access for top-ranked proposals is guaranteed by the f1=1 rule (the first period is fully scheduled) and the c=3 month removal rule caps how long proposals can wait, so the 'average two months' is partly an artifact of the chosen parameters rather than an independent prediction. The paper also never simulates the conventional call-based model in the same framework, so there is no controlled comparison to back the 'nine months vs. two months' contrast.\n\nNone of this kills the proposal. The scheduling concept is coherent, and the test phase at PETRA III can provide real validation. But the paper currently overstates what the simulation demonstrates. For someone who cares about synchrotron user operations, this is worth engaging with; for a general physics audience, it is a niche operations read.\n\nRecommendation: send it to peer review, but the referees should insist on resolving the 450/110 issue, publishing a full algorithmic specification or code, and either reporting test-phase data or softening the abstract's quantitative claims until then.","headline":"Sensible rolling-access proposal from DESY, but the wait-time claims are only as strong as a simulation that doesn't add up as reported.","tokens_in":9245,"tokens_out":2126,"would_cite":false,"duration_ms":26274,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"deepseek-v4-flash","headline":"A generic rolling access model can reduce synchrotron beamtime waits from about nine months to about two months while keeping expert review.","keywords":["synchrotron radiation","beamtime access","rolling access model","proposal scheduling","dynamic ranking","scheduling simulation","user facilities"],"falsifier":"Measure the actual submission-to-beamtime distribution from the five test beamlines over a full year: if top-ranked academic proposals consistently wait more than 40 days, or the average wait is not near two months, the model's central quantitative claim is contradicted.","tokens_in":1545,"feed_emoji":"🔬","tokens_out":2302,"duration_ms":101232,"temperature":0.7,"pith_summary":"The paper argues that the standard bi-annual and tri-annual proposal cycles used by synchrotron light sources can be replaced by a single rolling process in which proposals arrive, are reviewed, and are scheduled continuously. Its central quantitative claim is that this generic rolling access model cuts the wait between proposal submission and beamtime to about two months on average, with the strongest proposals scheduled within 15 to 40 days, compared with roughly nine months under the conventional call-based system. The paper supports this with a scheduling simulation of 450 virtual proposals over 730 days, using normally distributed ratings and requested shifts between 3 and 18, and describes a live test on five beamlines spanning diffraction, spectroscopy, and crystallography. The payoff, if the model works, is faster and more flexible access, a single deadline-free submission route, and the ability to spread approved shifts over multiple beamtime blocks without weakening external review.","feed_headline":"Rolling access model cuts synchrotron waits to ~2 months","feed_subtitle":"A year-round proposal and scheduling scheme puts top-ranked experiments on the beam in 15–40 days.","key_machinery":"The load-bearing mechanism is a continuously updated proposal stack combined with a look-ahead schedule divided into n time slices, each with its own filling factor f_i, meaning the fraction of available shifts already booked. New proposals are ranked by an absolute rating from external reviewers, refined in regular panel meetings, and inserted into the stack immediately, so the ranking always reflects the current set of requests. The schedule is only partially filled in the near future—for example, the first slice may be fully booked while later slices are deliberately left at 30% and 10%—so that highly ranked urgent proposals can be fitted quickly while long-preparation projects are schedu","core_discovery":"This paper claims that the conventional call-based access model used by most synchrotron facilities—where proposals are collected at fixed deadlines, reviewed together, and then scheduled—can be replaced by a generic rolling access model that handles submission, evaluation, and scheduling continuously. The model's quantitative promise is a reduction in the typical wait from proposal submission to beamtime from roughly nine months under bi-annual calls to about two months on average, with the highest-ranked proposals scheduled within 15 to 40 days, and about two weeks for commercial customers at some beamlines. The evidence is a simulation of 450 virtual proposals arriving uniformly over 730","pith_inferences":["The paper leaves implicit that the same stack-and-filling-factor machinery could apply to other proposal-driven shared research facilities, such as neutron sources, free-electron lasers, or central imaging laboratories.","A testable consequence the paper does not pursue is inserting fixed 'benchmark' proposals at regular intervals to check whether absolute reviewer ratings drift as the proposal pool changes over time.","Because the simulation excludes review time and uses synthetic arrival and rating distributions, the natural validation is a before/after comparison of real submission-to-beamtime distributions once enough test-phase data accumulate.","The two simulation runs reveal a tunable trade-off between access speed and beamtime utilization—more aggressive scheduling gives faster access but more unscheduled days—so each beamline could optimize its own slice lengths and filling factors."],"forward_implications":["Proposal submission becomes deadline-free: one unified scheme replaces separate calls for regular, long-term, block-allocation, and rapid-access proposals, with only two templates split by requested shifts.","The average wait between submission and first beamtime falls from about nine months to roughly two months for academic users, and to about two weeks for commercial customers on tested beamlines.","Users can split approved shifts across multiple beamtime blocks and, once project-based access is added, can request multiple beamlines through a single proposal, enabling milestone-driven experiments.","Proposals stay valid longer (up to two years in the test), giving teams flexible preparation time beyond a single scheduling period without losing priority.","The scheduling parameters can be tuned per beamline, including a parameter set that reproduces the conventional six-month call model, so facilities can migrate gradually rather than switching abruptly."],"supporting_citations":[{"why":"Documents existing adaptations and rolling or multiple-call access modes at other facilities, supplying the baseline the generic model is meant to extend and improve.","marker":"[10]"},{"why":"Describes the multi-setup high-resolution diffraction beamline included in the test rollout, supporting the claim that the model covers beamlines with several instruments.","marker":"[11]"},{"why":"Describes the hard X-ray photoelectron spectroscopy beamline with multiple vacuum and pressure setups, providing a second diverse test case.","marker":"[14]"},{"why":"Describes the macromolecular crystallography beamline and its cryogenic in-vacuum microscope, a test case for high-throughput automated beamlines.","marker":"[16]"},{"why":"Summarizes the status of crystallography beamlines at the facility, giving context for the two crystallography beamlines in the test.","marker":"[17]"},{"why":"Presents the planned ultralow-emittance successor source, the future deployment target that motivates the need for a generic access model.","marker":"[18]"}],"fun_headline_variants":["Rolling access slashes synchrotron wait from 9 months to 2","Top synchrotron proposals get beamtime in 15–40 days with rolling access","Generic rolling access model replaces call-based waits at synchrotrons","Rolling access: year-round proposal submission cuts synchrotron wait"],"cache_read_input_tokens":10752,"weakest_assumption_plain":"The headline wait times rest on a simulation whose inputs—uniform random proposal arrivals, normally distributed ratings, 3 to 18 shift requests, and a three-month removal rule—are assumed rather than measured from real facility operations.","fun_headline_variants_meta":{"raw":{"variants":["Rolling access slashes synchrotron wait from 9 months to 2","Top synchrotron proposals get beamtime in 15–40 days with rolling access","Generic rolling access model replaces call-based waits at synchrotrons","Rolling access: year-round proposal submission cuts synchrotron wait"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000952,"raw_usage":{"total_tokens":3828,"prompt_tokens":602,"completion_tokens":3226,"prompt_tokens_details":{"cached_tokens":256},"prompt_cache_hit_tokens":256,"prompt_cache_miss_tokens":346,"completion_tokens_details":{"reasoning_tokens":3140}},"tokens_in":346,"tokens_out":3226,"duration_ms":23188,"temperature":1.0,"reasoning_tokens":3140,"cache_read_input_tokens":256,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-05T11:42:47.484089+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Measure the actual submission-to-beamtime distribution from the five test beamlines over a full year: if top-ranked academic proposals consistently wait more than 40 days, or the average wait is not near two months, the model's central quantitative claim is contradicted.","supporting_citations":[{"cited_title":"ESRF Prepares New User Access Modes","cited_arxiv_id":null,"evidence_quote":"Documents existing adaptations and rolling or multiple-call access modes at other facilities, supplying the baseline the generic model is meant to extend and improve."},{"cited_title":"The high-resolution diffraction beamline P08 at PETRA III","cited_arxiv_id":null,"evidence_quote":"Describes the multi-setup high-resolution diffraction beamline included in the test rollout, supporting the claim that the model covers beamlines with several instruments."},{"cited_title":"The new dedicated HAXPES beamline P22 at PETRA III","cited_arxiv_id":null,"evidence_quote":"Describes the hard X-ray photoelectron spectroscopy beamline with multiple vacuum and pressure setups, providing a second diverse test case."},{"cited_title":"Development of an in-vacuum x-ray microscope with cryogenic sample cooling for beamline P11 at PETRA III","cited_arxiv_id":null,"evidence_quote":"Describes the macromolecular crystallography beamline and its cryogenic in-vacuum microscope, a test case for high-throughput automated beamlines."},{"cited_title":"Status of the crystallography beamlines at PETRA III","cited_arxiv_id":null,"evidence_quote":"Summarizes the status of crystallography beamlines at the facility, giving context for the two crystallography beamlines in the test."},{"cited_title":"PETRA IV: the ultralow-emittance source project at DESY","cited_arxiv_id":null,"evidence_quote":"Presents the planned ultralow-emittance successor source, the future deployment target that motivates the need for a generic access model."}],"review_version":1}