{"id":"f223d54a-6984-483d-a2ce-507f792b25c8","arxiv_id":"2502.00002","paper_version":1,"verdict":"CONDITIONAL","confidence":"HIGH","novelty_score":6.0,"correctness_risk":"low","formal_verification":"none","parameter_count":0,"one_line_summary":"A systematic review of 34 papers shows digital twins in tourism are an early-stage field, dominated by cultural heritage, destination management, and one-way data links.","lead":"This paper systematically reviews 34 studies on digital twin technology applied to tourism, finding that the field is new and mostly focused on digitizing cultural heritage. It is useful as a map of an emerging niche, showing what has been built, what data and interfaces are used, and where the open gaps are.","discovery_kind":"review","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The corpus-selection step is the load-bearing point: because the search requires the literal phrase 'digital twin' in title/abstract/keywords, functionally equivalent tourism systems described with other terms are excluded, and the 'early stage / research gap' conclusions could be an artifact of…","rationale":"I read the manuscript as a descriptive mapping SLR: its value is the first systematic characterization of how digital twin technology is used in tourism. The methodology is transparent (PRISMA flow, explicit taxonomy), and the reader's complaint about data-integrity errors in Figures 5 and 9 is accurate and worth correcting. However, I do not see those labeling errors as the most load-bearing point for the central claim; they affect a handful of bibliometric counts and are easily repaired. The more fundamental condition is that the corpus was assembled by requiring the literal term 'digital twin' in title/abstract/keywords. A real digital-twin tourism system reported as a 'smart heritage metaverse' or a 'virtual tourism platform' would be missed even if it met every inclusion criterion. This directly bears on RQ1 and RQ3, where the paper's headline conclusions ('oldest study 2021', small corpus, large research gap) live. The authors do acknowledge this limitation in Section 5.3.4, which is honest and a point in their favor, but an acknowledged threat is still a threat unless its magnitude is bounded. Because the manuscript is currently CONDITIONAL on data corrections, and the search-coverage question is a separate condition that should also be demonstrated, I would keep the verdict CONDITIONAL (no change from the reader). The concrete test is a sensitivity search; it is feasible with the same databases and would confirm or retire the concern.","tokens_in":18346,"tokens_out":7763,"duration_ms":75360,"concrete_test":"Run a sensitivity search in the same three databases using an expanded query that keeps the tourism filter ('touris*' OR tourism OR tourist) but broadens the twin-side terms to include 'digital twin' OR 'digital shadow' OR 'digital model' OR 'virtual replica' OR 'metaverse' OR 'virtual tourism' OR 'smart heritage', in title/abstract/keywords, limited to 2020–2024; apply the same inclusion/exclusion criteria from Section 4.5 and compare the resulting corpus with the 34 studies. A decisive result: if the expanded set adds more than about 10 qualifying studies, or if any qualifying study is dated before 2021, recompute the tourism-type, purpose, spatial-scale, and data-linkage distributions.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The SLR's central descriptive findings (RQ1: 'early stage', oldest study 2021, only 34 studies; RQ3: 'extensive research gap') depend on the assumption in Section 4.4 that all relevant studies contain the literal string 'digital twin' plus 'touris*' in the title, abstract, or keywords of Scopus, Web of Science, or Dimensions. Studies that describe functionally equivalent systems—for example, a 'virtual tourism platform', a 'smart heritage metaverse', a 'digital shadow' of a destination, or a 'virtual replica' with live sensor feedback—are invisible to this search even when they satisfy inclusion criterion IC3. The authors acknowledge this as a construct validity threat in Section 5.3.4 and mention snowballing as a possible enrichment, but no sensitivity analysis estimates how many relevant studies are missed or whether their absence changes the reported distributions. If a non-negligible set of synonym-described studies exists, then 'the oldest study dates to 2021' and 'the topic is at an early, evolving stage' may be statements about the query string rather than about the field. The Figure 5 'Italia'/'Italy' duplicate and Figure 9 'UCG'/'UGC' double-counting are real and should be fixed, but they affect specific counts; the query-coverage issue threatens the review's central claim.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"This paper conducts a systematic literature review (SLR) with bibliometric analysis to characterize the state of the art of digital twin technology in tourism. Using a search for the literal strings 'digital twin' and 'touris*' in three databases, the authors select 34 peer-reviewed studies and classify them along a taxonomy of tourism type, purpose, spatial scale, data sources, data linkage, visualization, and application. The reported findings are that the field is nascent (oldest study from 2021), that cultural tourism and heritage digitalization dominate, that destination management is the primary purpose, that spatial scales are mostly site-level, that data linkage is often unilateral, and that there is a large research gap.","tokens_in":18537,"tokens_out":5977,"duration_ms":50511,"significance":"If the findings are robust, this SLR offers a useful and timely mapping of an emerging interdisciplinary area, identifies concrete research gaps, and provides a reproducible classification scheme. The paper is transparent in its PRISMA-based methodology and includes an explicit validity-threat discussion. However, the significance is conditional on the adequacy of the search scope—since functionally equivalent systems described with other terms may be missed—and on the correction of several data tabulation errors that currently undermine confidence in the bibliometric results.","major_comments":[{"comment":"The search string requires the literal phrase 'digital twin' together with 'touris*' in the title, abstract, or keywords. This excludes studies describing functionally equivalent systems with alternative terminology, such as 'virtual replica', 'digital shadow', or 'smart heritage metaverse', even when they satisfy inclusion criterion IC3. The authors acknowledge this as a construct validity threat in Section 5.3.4, but they do not quantify its impact or provide a sensitivity analysis. Because the central claims of RQ1 ('early stage', oldest study 2021, only 34 studies) and RQ3 ('extensive research gap') are directly based on the size and composition of the corpus, the conclusions may reflect the query string rather than the actual state of the field. I recommend either broadening the search to include commonly used synonyms or performing a sensitivity analysis to demonstrate that the main findings are robust to terminological variation.","section":"Section 4.4 and Section 5.3.4"},{"comment":"The country frequency chart in Figure 5 lists 'Italia' and 'Italy' as separate entries with counts of 2 and 3 respectively, while the prose in Section 5.2.1 states that the United Kingdom and the Netherlands are the second most contributing countries with N=4 each. If the Italian entries are combined (total 5), Italy would be second, contradicting the text. This data integrity error should be corrected, and the corresponding prose and figure should be reconciled.","section":"Section 5.2.1 and Figure 5"},{"comment":"The data source frequency chart in Figure 9 lists both 'UCG' and 'UGC' as separate categories, with counts of 1 and 3 respectively. The taxonomy in Section 4.7.5 uses the abbreviation 'UGC' for user-generated data, and the text in Section 5.2.4 states that only N=3 studies use this source. The duplicate label appears to misrepresent the same category and leads to an inconsistent count. Please harmonize the labels and ensure the figure matches the text and taxonomy.","section":"Section 5.2.4 and Figure 9"}],"minor_comments":[{"comment":"The text states that the related studies 'start in 2020', whereas the abstract and Figure 3 indicate that the oldest study is from 2021. This inconsistency should be resolved.","section":"Section 5.3.1"},{"comment":"The abbreviation for user-generated data is inconsistently written as 'UCG' instead of 'UGC' in several places, including the taxonomy description and Figure 9. Please correct the abbreviation throughout.","section":"Section 4.7.5 and Figure 9"},{"comment":"The stacked bar chart for publication year and venue type is difficult to read because the segment values are not clearly labelled. Adding a legend or explicit numeric labels for each segment would improve clarity.","section":"Figure 3"}],"recommendation":"major_revision","confidential_remarks":null},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Worth a look if you work on tourism informatics or on digital twins outside manufacturing. It is the first SLR specifically on digital twins in tourism, and that alone earns it a read. The authors are transparent about their search and selection process, and their taxonomy (tourism type, purpose, spatial scale, data sources, data linkage, visualization, application) is a serviceable frame for grouping 34 studies. The main descriptive findings -- work starts around 2021, cultural heritage dominates, most systems are site-level, and half of the papers describe only unilateral physical-to-digital data flow -- are direct observations from the corpus and hold up. The summary table classifying each study is a useful artifact.\n\nThe soft spots are real but not fatal. The bibliometric figures contain clear data-integrity errors: Italy is split into 'Italia' and 'Italy' in Figure 5, which inflates its count; 'UCG' and 'UGC' both appear in Figure 9, double-counting user-generated content; and 'Traffic data' shows up in Figure 9 without being defined in the taxonomy. These need correction before anyone trusts the specific counts. The larger limitation is the search string itself: requiring the literal phrase 'digital twin' plus 'touris*' excludes papers that describe functionally equivalent systems as 'virtual tourism platforms' or 'smart heritage metaverse'. The authors acknowledge this in Section 5.3.4, but they don't estimate how many studies are missed or whether the 'extensive research gap' conclusion would survive a broader search. That caveat is worth stating as a boundary on the claims, though it is not disqualifying—the review's scope is the term as used, and the method is otherwise solid.\n\nThe circularity concern is minor. The taxonomy is derived from the corpus, but the headline findings (recent origin, cultural tourism focus, unilateral linkage) stand apart from the taxonomy and would survive even if the classification categories changed.\n\nWho gets value? Researchers wanting a structured entry point into digital twin work in tourism, and destination managers looking for a landscape of what has been tried. It is a mapping, not a technical breakthrough. That is fine. I would send it to a serious referee, with a request to fix the data errors and add a paragraph on the synonym-search sensitivity. My own verdict would be conditional acceptance for those revisions.","headline":"Useful first systematic map of digital twins in tourism, but the bibliometric tabulation has fixable errors and the search-string limitation means 'early stage' describes the term, not necessarily the underlying ideas.","tokens_in":19148,"tokens_out":1989,"would_cite":true,"duration_ms":21442,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"deepseek-v4-flash","headline":"Tourism's digital twins are young, heritage-focused, and often one-way, a systematic review of 34 studies finds.","keywords":["digital twin","tourism","systematic literature review","cultural heritage","destination management","data linkage","smart tourism","virtual tourism"],"falsifier":"A replication with a broader search—adding terms such as 'metaverse', 'virtual tourism', 'heritage information modeling', 'smart destination', and grey literature—would settle the claim. If that search uncovered tourism twin studies before 2021, or found a majority of systems with automatic digital-to-physical control, the paper's central picture of an early, mostly unilateral field would not hold.","tokens_in":18097,"feed_emoji":"🏛️","tokens_out":6614,"duration_ms":59706,"temperature":0.7,"pith_summary":"This paper asks how far digital twin technology has spread into tourism and answers with a systematic review of 34 peer-reviewed studies. It finds the field is at an early, evolving stage: the oldest study dates to 2021, cultural heritage dominates, destination management is the leading purpose, and most twins operate at site or local scale. A central observation is that half the reviewed systems move data only from the physical world to the digital model, so they do not meet the usually accepted requirement of a bilateral twin link. The review also finds more applied prototypes than theoretical frameworks, suggesting the technology is moving into practice faster than its conceptual foundations. If right, this gives researchers and destination managers a map of where digital twinning in tourism stands and where the gaps are.","feed_headline":"Most tourism 'digital twins' are one-way models, review finds","feed_subtitle":"Review of 34 studies since 2021 finds heritage-focused twins that rarely feed data back into the real world.","key_machinery":"The carrier of the argument is a seven-criterion classification taxonomy—tourism type, purpose, spatial scale, data sources, data linkage, visualization, and application—applied to each reviewed study. Within that taxonomy, the data-linkage criterion (unilateral versus bilateral) does the decisive work: it is the criterion that separates 'digital representation' from a twin in the strong sense, and it drives the paper's main gap claim. The taxonomy turns a set of heterogeneous case studies into comparable categories that support the bibliometric and thematic findings.","core_discovery":"The paper's central claim is that the state of the art in tourism digital twinning is immature and uneven. Most of the 34 studies concern cultural tourism, especially digitising heritage buildings, monuments, and archaeological sites, and their main purpose is destination management. Spatially, most systems represent individual sites or localities, and the most common data sources are 3D scanning and GIS data. The decisive finding is about connectivity: in half the studies the data linkage is unilateral, and in many of the studies classified as bilateral the digital-to-physical path is indirect, so the physical counterpart changes only when people act on the twin's output rather than through automated synchronisation. The review interprets this as a mismatch with mainstream DT definitions and identifies fuller bilateral linkage, dynamic data sources, and larger spatial scales as the main open directions.","pith_inferences":["Editorial inference: the unilateral/bilateral split could be used as a maturity ladder for tourism twins, letting future studies grade systems by whether the digital-to-physical path is passive, advisory, or automatic.","Editorial inference: if bilateral, data-rich tourism twins become common, data governance and tourist privacy will become central design constraints rather than side issues; the review mentions governance only through one included framework.","Editorial inference: the one-way representation of heritage sites may also be a sampling artifact of funding and scanning costs rather than a stable property of the domain; cheaper reality capture could change the pattern quickly."],"forward_implications":["If the field is as young as the review suggests, the absence of studies before 2021 means there is no established baseline or best-practice pattern yet for tourism twins.","Because many tourism twins are unilateral, a reader should expect that systems labelled 'digital twins' in tourism often function as monitoring or visualisation tools, not closed-loop control systems.","The concentration on cultural heritage and site-level twins means the main near-term gains are likely in heritage preservation and site management, not in city-wide or national tourism governance.","The review's gap list points to concrete next steps: add dynamic data sources such as sensors, user-generated content, and web mining, and build bilateral digital-to-physical feedback for predicting crowding and service quality."],"supporting_citations":[{"why":"Supplies the validated systematic review methodology and step-by-step protocol the paper follows.","marker":"Carrera-Rivera et al. (2022)"},{"why":"Underpins the PRISMA flow used to report the identification, screening, and selection of the 34 studies.","marker":"Page et al. (2021)"},{"why":"Supplies the DT archetype definitions and the bilateral-twinning standard used to judge data linkage.","marker":"van der Valk et al. (2022)"},{"why":"Gives the compact definition of a digital twin as a virtual representation of a connected physical asset.","marker":"AIAA Digital Engineering Integration Committee (2020)"},{"why":"Defines the twinning-connection characteristics that shape the review's unilateral-versus-bilateral data-linkage criterion.","marker":"Jones et al. (2020)"},{"why":"The closest existing review, focused on digital twins for world-heritage sites in China, which the paper positions its tourism-specific review against.","marker":"Dang et al. (2023)"}],"fun_headline_variants":["Tourism digital twins rarely sync back, review finds","Study: Half of tourism twins are one-way data streams","Digital twins in tourism: mostly one-way, review says","Tourism twin tech lags industry definition, review finds","Review: Tourism twins lack true two-way connectivity"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The review's completeness rests on the assumption that searching three databases for the literal terms 'digital twin' and 'touris*' in title, abstract, and keywords captures all relevant studies; relevant work described as 'virtual tourism', 'smart heritage metaverse', or 'city digital replica' without those exact terms would be missed and could change the reported gaps, a limitation the authors acknowledge in Section 5.3.4.","fun_headline_variants_meta":{"raw":{"variants":["Tourism digital twins rarely sync back, review finds","Study: Half of tourism twins are one-way data streams","Digital twins in tourism: mostly one-way, review says","Tourism twin tech lags industry definition, review finds","Review: Tourism twins lack true two-way connectivity"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000157,"raw_usage":{"total_tokens":1244,"prompt_tokens":989,"completion_tokens":255,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":605,"completion_tokens_details":{"reasoning_tokens":177}},"tokens_in":605,"tokens_out":255,"duration_ms":3324,"temperature":1.0,"reasoning_tokens":177,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-10T22:21:09.084914+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"A replication with a broader search—adding terms such as 'metaverse', 'virtual tourism', 'heritage information modeling', 'smart destination', and grey literature—would settle the claim. If that search uncovered tourism twin studies before 2021, or found a majority of systems with automatic digital-to-physical control, the paper's central picture of an early, mostly unilateral field would not hold.","supporting_citations":[{"cited_title":"URL: https://www.aia-aerospace.org/wp-content/uploads/Digital-Twin-Institute-Position-Paper-December-2020-1.pdf B","cited_arxiv_id":null,"evidence_quote":"Gives the compact definition of a digital twin as a virtual representation of a connected physical asset."}],"review_version":1}