{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,7,31]],"date-time":"2026-07-31T08:59:47Z","timestamp":1785488387088,"version":"3.56.0"},"reference-count":35,"publisher":"PeerJ","license":[{"start":{"date-parts":[[2026,7,31]],"date-time":"2026-07-31T00:00:00Z","timestamp":1785456000000},"content-version":"unspecified","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"name":"Research on Streaming Processing Method for Massive Spatiotemporal Trajectory Data in New Smart Cities","award":["KJQN202301905"],"award-info":[{"award-number":["KJQN202301905"]}]},{"name":"Research on Key Technologies for Stream Processing of Spatiotemporal Trajectory Data in the Context of Digital Economy","award":["KJZD-K202401902"],"award-info":[{"award-number":["KJZD-K202401902"]}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"abstract":"<jats:p>\n                    Accurate traffic flow prediction is the core foundation of Intelligent Transportation Systems (ITS) and urban traffic dynamic optimization. However, existing mainstream prediction models have three critical limitations: poor cross-road transferability, mandatory separate training for individual road sections, and insufficient mining of historical traffic peak information, which jointly restrict prediction accuracy and engineering scalability in practical urban traffic management. To address these challenges, this article proposes the Multi-road Multi-Peak-aware Informer (MMP-Informer), an improved Informer-based long-term time series prediction model with a novel dual-peak-aware attention mechanism. This core module adaptively locates historical maximum and minimum traffic peaks\n                    <jats:italic>via<\/jats:italic>\n                    an adaptive window matched to input sequence length, introduces learnable parameters to weight the influence of different peaks on future traffic flow, and realizes single-model end-to-end simultaneous multi-road traffic flow prediction. Meanwhile, the model retains the encoder-decoder architecture of the original Informer, integrates self-attention distillation in the encoder to reduce computational overhead, and adopts a generative decoder paradigm to avoid error accumulation in multi-step prediction. Experiments on the 2018 University of Technology Sydney (UTS) multi-source traffic dataset show that MMP-Informer consistently outperforms state-of-the-art baselines including SCINet, NsTransformer, Transformer, and original Informer across 24-, 48-, and 96-h prediction horizons. For 24-h short-term prediction, it achieves a Mean Absolute Error (MAE) of 25.43 and Mean Absolute Percentage Error (MAPE) of 11.55%; for 96-h long-term prediction, it reaches an MAE of 31.21 and MAPE of 12.93%, where the MAE is 2.51 lower and MAPE is 1.97 percentage points lower than the original Informer, respectively. Ablation tests verify that the dual-peak-aware design significantly enhances the model\u2019s adaptability to both low-traffic and high-traffic scenarios, balancing prediction accuracy and long-term forecasting stability. Supplementary experiments on merged Electricity Transformer Temperature hourly (ETTh1\/ETTh2) datasets further confirm its stable and superior performance in non-traffic time series scenarios, with strong cross-scenario generalization ability. The proposed model eliminates the high cost of single-road separate training, breaks through the cross-road transferability bottleneck of traditional models, maintains stable accuracy in both short-term and long-term prediction, and has high practical engineering value for large-scale deployment in urban intelligent traffic management.\n                  <\/jats:p>","DOI":"10.7717\/peerj-cs.4035","type":"journal-article","created":{"date-parts":[[2026,7,31]],"date-time":"2026-07-31T08:30:20Z","timestamp":1785486620000},"page":"e4035","source":"Crossref","is-referenced-by-count":0,"title":["MMP-Informer: an informer-based traffic flow prediction model by leveraging the previous cycle\u2019s max-min peaks"],"prefix":"10.7717","volume":"12","author":[{"given":"Fei","family":"Meng","sequence":"first","affiliation":[{"name":"College of Computer Science, ChongQing University, Chongqing, China"},{"name":"College of Big Data and Artificial Intelligence, Chongqing Institute of Engineering, Chongqing, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"SuBo","family":"Zhao","sequence":"additional","affiliation":[{"name":"College of Computer Science, ChongQing University, Chongqing, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Qing","family":"Zhou","sequence":"additional","affiliation":[{"name":"College of Computer Science, ChongQing University, Chongqing, China"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"4443","published-online":{"date-parts":[[2026,7,31]]},"reference":[{"issue":"18","key":"10.7717\/peerj-cs.4035\/ref-1","doi-asserted-by":"publisher","first-page":"20722","DOI":"10.1109\/JSEN.2021.3100324","article-title":"Using noise pollution data for traffic prediction in smart cities: experiments based on LSTM recurrent neural networks","volume":"21","author":"Awan","year":"2021","journal-title":"IEEE Sensors Journal"},{"issue":"3","key":"10.7717\/peerj-cs.4035\/ref-2","doi-asserted-by":"publisher","first-page":"736","DOI":"10.1111\/tgis.12644","article-title":"Traffic transformer: capturing the continuity and periodicity of time series for traffic forecasting","volume":"24","author":"Cai","year":"2020","journal-title":"Transactions in GIS"},{"key":"10.7717\/peerj-cs.4035\/ref-3","first-page":"103","article-title":"On the properties of neural machine translation: encoder\u2013decoder approaches","author":"Cho","year":"2014"},{"issue":"11","key":"10.7717\/peerj-cs.4035\/ref-4","doi-asserted-by":"publisher","first-page":"4883","DOI":"10.1109\/TITS.2019.2950416","article-title":"Traffic graph convolutional recurrent neural network: a deep learning framework for network-scale traffic learning and forecasting","volume":"21","author":"Cui","year":"2020","journal-title":"IEEE Transactions on Intelligent Transportation Systems"},{"key":"10.7717\/peerj-cs.4035\/ref-5","doi-asserted-by":"publisher","first-page":"143025\u2013143035","DOI":"10.1109\/access.2019.2941280","article-title":"Short-term traffic flow prediction method for urban road sections based on spacetime analysis and GRU","volume":"7","author":"Dai","year":"2019","journal-title":"IEEE Access"},{"key":"10.7717\/peerj-cs.4035\/ref-6","doi-asserted-by":"publisher","first-page":"95958","DOI":"10.1109\/access.2023.3311818","article-title":"Spatiotemporal exogenous variables enhanced model for traffic flow prediction","volume":"11","author":"Dong","year":"2023","journal-title":"IEEE Access"},{"issue":"4, SI","key":"10.7717\/peerj-cs.4035\/ref-7","first-page":"1345","article-title":"A deep-learning model for urban traffic flow prediction with traffic events mined from twitter","volume":"24","author":"Essien","year":"2021","journal-title":"World Wide Web-Internet and Web Information Systems"},{"issue":"6","key":"10.7717\/peerj-cs.4035\/ref-8","doi-asserted-by":"publisher","first-page":"2154","DOI":"10.1109\/tmc.2020.2976936","article-title":"BuildSenSys: reusing building sensing data for traffic prediction with cross-domain learning","volume":"20","author":"Fan","year":"2021","journal-title":"IEEE Transactions on Mobile Computing"},{"key":"10.7717\/peerj-cs.4035\/ref-9","doi-asserted-by":"publisher","first-page":"4795","DOI":"10.1109\/access.2018.2888561","article-title":"STANN: a spatio-temporal attentive neural network for traffic prediction","volume":"7","author":"He","year":"2019","journal-title":"IEEE Access"},{"issue":"8","key":"10.7717\/peerj-cs.4035\/ref-10","doi-asserted-by":"publisher","first-page":"1735","DOI":"10.1162\/neco.1997.9.8.1735","article-title":"Long short-term memory","volume":"9","author":"Hochreiter","year":"1997","journal-title":"Neural Computation"},{"issue":"15","key":"10.7717\/peerj-cs.4035\/ref-11","doi-asserted-by":"publisher","first-page":"6733","DOI":"10.1016\/j.amc.2011.01.073","article-title":"Hybrid evolutionary algorithms in a SVR traffic flow forecasting model","volume":"217","author":"Hong","year":"2011","journal-title":"Applied Mathematics and Computation"},{"key":"10.7717\/peerj-cs.4035\/ref-12","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.1412.6980","article-title":"Adam: a method for stochastic optimization","author":"Kingma","year":"2017"},{"key":"10.7717\/peerj-cs.4035\/ref-13","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.1609.02907","article-title":"Semi-supervised classification with graph convolutional networks","author":"Kipf","year":"2016"},{"issue":"11","key":"10.7717\/peerj-cs.4035\/ref-14","doi-asserted-by":"publisher","first-page":"2278","DOI":"10.1109\/5.726791","article-title":"Gradient-based learning applied to document recognition","volume":"86","author":"LeCun","year":"1998","journal-title":"Proceedings of the IEEE"},{"key":"10.7717\/peerj-cs.4035\/ref-15","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.2310.06625","article-title":"iTransformer: inverted transformers are effective for time series forecasting","author":"Liu","year":"2023b"},{"issue":"7","key":"10.7717\/peerj-cs.4035\/ref-16","doi-asserted-by":"publisher","first-page":"3866","DOI":"10.3390\/app15073866","article-title":"A review of traffic flow prediction methods in intelligent transportation system construction","volume":"15","author":"Liu","year":"2025","journal-title":"Applied Sciences"},{"issue":"34","key":"10.7717\/peerj-cs.4035\/ref-17","doi-asserted-by":"publisher","first-page":"24207","DOI":"10.1007\/s00521-023-09003-z","article-title":"Multivariate long-time series traffic passenger flow prediction using causal convolutional sparse self-attention MTS-informer","volume":"35","author":"Liu","year":"2023a","journal-title":"Neural Computing & Applications"},{"key":"10.7717\/peerj-cs.4035\/ref-18","doi-asserted-by":"crossref","DOI":"10.52202\/068431-0718","article-title":"Non-stationary transformers: exploring the stationarity in time series forecasting","volume-title":"Advances in Neural Information Processing Systems 35 (NEURIPS 2022). 36th Conference on Neural Information Processing Systems (NeurIPS), Electr Network, Nov 28\u2013Dec 09, 2022","author":"Liu","year":"2022b"},{"key":"10.7717\/peerj-cs.4035\/ref-19","doi-asserted-by":"crossref","DOI":"10.52202\/068431-0421","article-title":"SCINet: time series modeling and forecasting with sample convolution and interaction","author":"Liu","year":"2022a"},{"issue":"2","key":"10.7717\/peerj-cs.4035\/ref-20","doi-asserted-by":"publisher","first-page":"865","DOI":"10.1109\/TITS.2014.2345663","article-title":"Traffic flow prediction with big data: a deep learning approach","volume":"16","author":"Lv","year":"2015","journal-title":"IEEE Transactions on Intelligent Transportation Systems"},{"issue":"4","key":"10.7717\/peerj-cs.4035\/ref-21","doi-asserted-by":"publisher","first-page":"818","DOI":"10.3390\/s17040818","article-title":"Learning traffic as images: a deep convolutional neural network for large-scale transportation network speed prediction","volume":"17","author":"Ma","year":"2017","journal-title":"Sensors"},{"issue":"6","key":"10.7717\/peerj-cs.4035\/ref-22","doi-asserted-by":"publisher","first-page":"2278","DOI":"10.1109\/TITS.2018.2867042","article-title":"Parallel architecture of convolutional bi-directional LSTM neural networks for network-wide metro ridership prediction","volume":"20","author":"Ma","year":"2019","journal-title":"IEEE Transactions on Intelligent Transportation Systems"},{"key":"10.7717\/peerj-cs.4035\/ref-23","doi-asserted-by":"publisher","first-page":"98053","DOI":"10.1109\/access.2019.2929692","article-title":"T-LSTM: a long short-term memory neural network enhanced by temporal information for traffic flow prediction","volume":"7","author":"Mou","year":"2019","journal-title":"IEEE Access"},{"key":"10.7717\/peerj-cs.4035\/ref-24","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.2211.14730","article-title":"A time series is worth 64 words: long-term forecasting with transformers","author":"Nie","year":"2023"},{"issue":"22","key":"10.7717\/peerj-cs.4035\/ref-25","doi-asserted-by":"publisher","first-page":"9024","DOI":"10.3390\/s22229024","article-title":"SGGformer: shifted graph convolutional graph-transformer for traffic prediction","volume":"22","author":"Pu","year":"2022","journal-title":"Sensors"},{"key":"10.7717\/peerj-cs.4035\/ref-26","first-page":"147","article-title":"PLCount: a probabilistic fusion algorithm for accurately estimating occupancy from 3D camera counts","author":"Sangoboye","year":"2016"},{"issue":"24","key":"10.7717\/peerj-cs.4035\/ref-27","doi-asserted-by":"publisher","first-page":"30148","DOI":"10.1007\/s10489-023-05126-x","article-title":"Transformer network with decoupled spatial-temporal embedding for traffic flow forecasting","volume":"53","author":"Sun","year":"2023","journal-title":"Applied Intelligence"},{"key":"10.7717\/peerj-cs.4035\/ref-28","article-title":"Attention is all you need","volume-title":"Advances in Neural Information Processing Systems (NIPS 2017). 31st Annual Conference on Neural Information Processing Systems (NIPS), Long Beach, CA, Dec 04\u201309, 2017","volume":"30","author":"Vaswani","year":"2017"},{"key":"10.7717\/peerj-cs.4035\/ref-29","first-page":"1907","article-title":"Graph WaveNet for deep spatial-temporal graph modeling","volume-title":"Proceedings of the Twenty-Eighth International Joint Conference on Artificial Intelligence, Macao, Peoples R China, Aug 10\u201316, 2019","author":"Wu","year":"2019"},{"issue":"11","key":"10.7717\/peerj-cs.4035\/ref-30","doi-asserted-by":"publisher","first-page":"22386","DOI":"10.1109\/TITS.2021.3102983","article-title":"Learning dynamic and hierarchical traffic spatiotemporal features with transformer","volume":"23","author":"Yan","year":"2022","journal-title":"IEEE Transactions on Intelligent Transportation Systems"},{"issue":"7","key":"10.7717\/peerj-cs.4035\/ref-31","doi-asserted-by":"publisher","first-page":"8276","DOI":"10.1109\/jiot.2024.3501294","article-title":"TST-Trans: a transformer network for urban traffic flow prediction","volume":"12","author":"Zhang","year":"2025","journal-title":"IEEE Internet of Things Journal"},{"key":"10.7717\/peerj-cs.4035\/ref-32","doi-asserted-by":"crossref","DOI":"10.1109\/ITSC45102.2020.9294470","article-title":"Attention based graph Bi-LSTM networks for traffic forecasting","author":"Zhao","year":"2020"},{"issue":"11","key":"10.7717\/peerj-cs.4035\/ref-33","doi-asserted-by":"publisher","first-page":"6910","DOI":"10.1109\/TITS.2020.2997352","article-title":"A hybrid deep learning model with attention-based Conv-LSTM networks for short-term traffic flow prediction","volume":"22","author":"Zheng","year":"2021","journal-title":"IEEE Transactions on Intelligent Transportation Systems"},{"key":"10.7717\/peerj-cs.4035\/ref-34","first-page":"11106","article-title":"Informer: beyond efficient transformer for long sequence time-series forecasting","volume":"35","author":"Zhou","year":"2021"},{"issue":"7","key":"10.7717\/peerj-cs.4035\/ref-35","doi-asserted-by":"publisher","first-page":"8221","DOI":"10.1109\/tits.2024.3411638","article-title":"MT-STNet: a novel multi-task spatiotemporal network for highway traffic flow prediction","volume":"25","author":"Zou","year":"2024","journal-title":"IEEE Transactions on Intelligent Transportation Systems"}],"container-title":["PeerJ Computer Science"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/peerj.com\/articles\/cs-4035.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/peerj.com\/articles\/cs-4035.xml","content-type":"application\/xml","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/peerj.com\/articles\/cs-4035.html","content-type":"text\/html","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/peerj.com\/articles\/cs-4035.pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,7,31]],"date-time":"2026-07-31T08:30:24Z","timestamp":1785486624000},"score":1,"resource":{"primary":{"URL":"https:\/\/peerj.com\/articles\/cs-4035"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,7,31]]},"references-count":35,"alternative-id":["10.7717\/peerj-cs.4035"],"URL":"https:\/\/doi.org\/10.7717\/peerj-cs.4035","archive":["CLOCKSS","LOCKSS","Portico"],"relation":{},"ISSN":["2376-5992"],"issn-type":[{"value":"2376-5992","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,7,31]]},"article-number":"e4035"}}