A Meta-Learning Algorithm for Rebalancing the Bike-Sharing System in IoT Smart City

Zhang, Cong; Wu, Fan; Wang, He; Tang, Bihua; Fan, Wenhao; Liu, Yuanan

doi:10.1109/jiot.2022.3176145

Cited by 10 publications

(5 citation statements)

References 22 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…To aggregate q-RFL assessment values @ ij of alternative x i on all criteria c j (j = 1, 2, 3, 4, 5) into the overall assessment value @ i of the alternative x i (i = 1, 2, 3, 4) the q-RFLWA operator Eq (12) is employed and the overall assessment values of alternatives x i (i = 1, 2, 3, 4) are obtained as: @ 1 = ((s 5 , −0.0063), h0.6003, 0.4698i), @ 2 = ((s 4 , 0.1969), h0.5066, 0.4408i), @ 3 = ((s 3 , −0.3348), h0.5100, 0.4381i), @ 4 = ((s 2 , −0.1469), h0.4954, 0.4383i).…”

Section: Q-rfl Aggregation-based Methodsmentioning

confidence: 99%

“…Furthermore, to the best of our knowledge, prior research on BS scarcely considers the choice of recycling supplier. After reviewing and summarising, the BS investigation concentrates mainly on site selection [5][6][7], demand prediction [8][9][10], and rebalancing approach [11,12]. Mete et al [5] conducted a case study focused on the Gaziantep University campus to identify potential station locations for students.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

An integrated group decision-making method under q-rung orthopair fuzzy 2-tuple linguistic context with partial weight information

Abbas,

Ali,

Mashwani

et al. 2024

PLoS ONE

View full text Add to dashboard Cite

Considering the advantages of q-rung orthopair fuzzy 2-tuple linguistic set (q-RFLS), which includes both linguistic and numeric data to describe evaluations, this article aims to design a new decision-making methodology by integrating Vlsekriterijumska Optimizacija I Kompromisno Resenje (VIKOR) and qualitative flexible (QUALIFLEX) methods based on the revised aggregation operators to solve multiple criteria group decision making (MCGDM). To accomplish this, we first revise the extant operational laws of q-RFLSs to make up for their shortcomings. Based on novel operational laws, we develop q-rung orthopair fuzzy 2-tuple linguistic (q-RFL) weighted averaging and geometric operators and provide the corresponding results. Next, we develop a maximization deviation model to determine the criterion weights in the decision-making procedure, which accounts for partial weight unknown information. Then, the VIKOR and QUALIFLEX methodologies are combined, which can assess the concordance index of each ranking combination using group utility and individual maximum regret value of alternative and acquire the ranking result based on each permutation’s general concordance index values. Consequently, a case study is conducted to select the best bike-sharing recycling supplier utilizing the suggested VIKOR-QUALIFLEX MCGDM method, demonstrating the method’s applicability and availability. Finally, through sensitivity and comparative analysis, the validity and superiority of the proposed method are demonstrated.

show abstract

Section: Q-rfl Aggregation-based Methodsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

An integrated group decision-making method under q-rung orthopair fuzzy 2-tuple linguistic context with partial weight information

Abbas,

Ali,

Mashwani

et al. 2024

PLoS ONE

View full text Add to dashboard Cite

show abstract

“…Heuristic algorithms are commonly used because they have a unique advantage in solving the problem of rebalancing large-scale stations. The main heuristic algorithms currently in use are the iterated tabu search heuristic [23], hybrid large neighborhood search [24], greedy-genetic heuristic [14], three-step math heuristic [22], cluster-first-route-second heuristic [29], artificial bee colony algorithm [8], PILOT/VNS/GRASP hybrid heuristic [19], and meta-learning algorithm [30]. Despite the advancement in heuristic algorithms, however, their solution speed is still not suitable for the design of static rebalancing problems in bike-sharing systems at medium-to large-scales.…”

Section: Related Workmentioning

confidence: 99%

“…Heuristic algorithms are used for their computational speed and ability to generate multiple sets of solutions in a short time. They include genetic algorithms [22], meta-learning algorithms [30], and ant colony algorithms [3]. This study chose a genetic iterative approach to compare it with the two-stage robust optimization algorithm.…”

Section: ) Comparison Of Algorithm Performancementioning

confidence: 99%

Robust Optimization Model Based on a Static Rebalancing Design for Bike- Sharing Systems Affected by Demand Uncertainty

Yu,

Xu,

Shi

et al. 2024

IEEE Access

View full text Add to dashboard Cite

Demand uncertainty and variations in user rental-return behavior can lead to an uneven spatial distribution of bicycles, which forces bike-sharing operators to rebalance the bike-sharing system by relocating bicycles from overstocked to understocked stations. In addition, this may lead to increased costs for bike-sharing operators. This paper proposed a two-stage robust model based on spatio-temporal networks to solve the static rebalancing problem in a bike-sharing system affected by demand uncertainty. The model designed a static rebalancing scheme for bike-sharing at a strategic level. To solve the problem, the paper used a customized column-and-constraint generation algorithm. Finally, the effectiveness of the proposed model and algorithm was confirmed using real data from Shanghai, China, providing strategic decision support for the static rebalancing problem in actual bike-sharing systems.

show abstract

“…Data Pre-processing: The data pre-processing techniques [15] include removing the missing values using mean, median or most frequently values imputation; the next step is to identify the abnormal values that need to be removed from the dataset. Feature selection is a crucial step in the modelling process, as it involves identifying and removing duplicate, unnecessary, or noisy features.…”

Section: ) Data Collection and Data Pre-processingmentioning

confidence: 99%

AI-driven rental bicycle system: An Ensemble learning approach

Long,

Hoang,

Aponte

et al. 2022

Proceedings of the Eighth International Conference on Research in Intelligent Computing in Engineering

View full text Add to dashboard Cite

Bicycle sharing is a notable sustainable transporta-tion option for metropolitan regions and communities seeking to address environmental concerns, reduce traffic congestion, and combat air pollution while promoting public health and improving connections. There are already technologies to support this system, including typical mobile applications and kiosks strategically positioned at the bicycle station. Nevertheless, most proposed solutions cannot accurately forecast the demand for bicycle availability, efficiently redistribute bicycles, create routes to circumvent traffic congestion and conduct comprehensive user analysis. To address these challenges, a framework for an AI-enabled bicycle-sharing system has been presented to predict the count of bicycle rentals. To assess performance, four distinct ensemble-based models are implemented and tested using various statistical parameters.

show abstract

A Meta-Learning Algorithm for Rebalancing the Bike-Sharing System in IoT Smart City

Cited by 10 publications

References 22 publications

An integrated group decision-making method under q-rung orthopair fuzzy 2-tuple linguistic context with partial weight information

An integrated group decision-making method under q-rung orthopair fuzzy 2-tuple linguistic context with partial weight information

Robust Optimization Model Based on a Static Rebalancing Design for Bike- Sharing Systems Affected by Demand Uncertainty

AI-driven rental bicycle system: An Ensemble learning approach

Contact Info

Product

Resources

About