Control Decisions Research Articles

Equipment control algorithm for cooling plant based on graph theory and path planning

This paper presents a novel path planning-based start-stop control algorithm for equipment in cooling plants designed to address the limitations of traditional serial start-stop control logic. The algorithm employs graph theory and path planning techniques to abstract the cooling plant schematic into a directed graph, enabling the selection of appropriate control paths and facilitating effective start-stop control decisions for the equipment. A key feature is the dynamic updating of the graph after each operation, which enables adaptive control target adjustments and prevents control blocking issues. Validation with real case scenarios demonstrates the effectiveness of the algorithm in automatic start-stop control, identifying equipment deviations, and handling faulty equipment switching. The algorithm also has the ability to handle some incorrectly turned on equipment, shutting down redundant equipment in the cooling plant system that is not in efficient operation. This algorithm shows the potential as an underlying execution algorithm of an advanced optimization algorithm to accelerate the implementation of advanced optimization algorithms.

A Cross-Sectional Study of the Perception of Individualized Nursing Care Among Nurses in Acute Medical and Perioperative Settings

Background/Objectives: Individualized nursing care allows for systematic assessment and intervention; considers a patient’s preferences, values, and context; and contributes to a positive care trajectory. However, its operationalization has proven to be challenging. This research aimed to evaluate nurses’ perceptions of individualized care and analyze their relationship with sociodemographic variables. Methods: A cross-sectional study was conducted on 122 eligible and registered nurses at a Hospital Center, in the Ophthalmology (operating room and inpatient ward) service, the Cardiology service, the Internal Medicine service, and the Medical Emergency Unit, for adults/older adults in Portugal. The nursing version of the Individualized Care Scale (ICS-Nurse) was used for the assessment, including three sub-dimensions: clinical situation, personal life situation, and decisional control over care-related decisions. Cronbach’s alpha and principal component analysis were used for the data analysis. The STROBE checklist was used to report this study. Results: No statistically significant differences were found based on the age, gender, level of education, or years of professional experience of the nurses within the sub-dimensions of individualization. The nurses overall had a good perception of the importance of individualized care (4.06 ± 0.46 ICS-A-NURSE) but faced difficulties in its implementation during the last shift they worked (3.97 ± 0.49 ICS-B-NURSE). Conclusions: The items considered to be of greatest importance were the response to the physical and emotional needs arising from illness and assistance in decision-making through educational instructions. Aspects related to the personal lives of patients, such as family inclusion in an individual’s care plan, everyday habits, and previous experiences of hospitalization, received the lowest scores. Recognizing priority areas for improvement in the individualization of nursing care can contribute to developing training programs and policies that promote a holistic approach. Future studies should consider patient outcomes related to their needs for individualization.

Runoff simulation of the Kaidu River Basin based on the GR4J-6 and GR4J-6-LSTM models

Study regionThe Kaidu River Basin originates from the southern slope of the Tienshan Mountains in the Xinjiang Uygur Autonomous Region, China. Study focusAccurate runoff simulation and prediction significantly affect flood control, drought resilience, and water resource allocation decisions. This study establishes the GR4J-6 model (modèle du Génie Rural à 4 paramètres Journalier-6, including a snowmelt module) and integrates it with the LSTM (Long Short-Term Memory) model to construct the hybrid GR4J-6-LSTM model and enhance the simulation accuracy of snowmelt runoff. A case study is conducted in the Kaidu River Basin to demonstrate the applicability of these models in cold and arid regions. The accuracy of the GR4J-6, LSTM, and GR4J-6-LSTM models is evaluated using Nash-Sutcliffe efficiency (NSE), Kling-Gupta Efficiency (KGE), and Root Mean Squared Error (RMSE) metrics. In addition, the contributions of each feature variable in the models are analyzed using the SHapley Additive exPlanations (SHAP) method to enhance the reliability of the results. New hydrological insights for the regionThe GR4J-6 model demonstrated good applicability in the Kaidu River Basin, with NSE, KGE, and RMSE values of 0.69, 0.79, and 39.39 m3/s during the validation period, respectively. The hybrid model GR4J-6-LSTM exhibited the highest comprehensive accuracy among all the models, with NSE, KGE, and RMSE values of 0.84, 0.87, and 28.79 m3/s, respectively. In the LSTM model, temperature and precipitation were found to significantly influence the simulated runoff, indicating that higher temperature and precipitation lead to increased runoff. In the GR4J-6-LSTM model, Tmin (minimum temperature) and the hydrological feature variable Qsim exhibited a strong positive correlation with simulated runoff, as Tmin and Qsim increased, they promoted stronger flow production. This study provides a framework for runoff simulation in snowmelt river basins, offering a reference for projecting extreme hydrological events under climate change.

Asynchronous decentralized traffic signal coordinated control in urban road network

AbstractThis study introduces an asynchronous decentralized coordinated signal control (ADCSC) framework for multi‐agent traffic signal control in the urban road network. The controller at each intersection in the network optimizes its signal control decisions based on a prediction of the future traffic demand as an independent agent. The asynchronous framework decouples the entangled interdependence between decision‐making and state prediction among different agents in decentralized coordinated decision‐making problems, enabling agents to proceed with collaborative decision‐making without waiting for other agents’ decisions. Within the proposed ADCSC framework, each controller dynamically optimizes its signal timing strategy with a unique rolling horizon scheme. The scheme's individualized parameters for each controller are determined based on the vehicle travel time between the adjacent intersections, ensuring that controllers can make informed control decisions with accurate arrival flow information from upstream intersections. The signal optimization problem is formulated as a mixed integer linear program model, which adopts a flexible signal scheme without a fixed phase structure and sequence. Simulation results demonstrate that the proposed ADCSC strategy significantly outperforms the benchmark signal coordination methods in terms of average delay, travel speed, stop numbers, and energy consumption. Experimental analysis on computation time validates the applicability of the proposed optimization model for real‐time implementation. Sensitivity analysis on key parameters in the framework is conducted, offering insights for parameter selection in practice. Furthermore, the ADCSC framework is extended to a road network in Qinzhou City, China, with 45 signalized intersections, demonstrating its effectiveness and scalability in the real‐world road network.

Fuzzy Method for Preliminary Classification of the Micro-Electromechanical Systems Data

Data of micro-electromechanical systems (sensors) are widely used in modern multipurpose intellectual environments, autonomous mobile robots and automated systems possessing intellectual components. Since the amount of data delivered by the sensors is rather considerable and their significance varies we propose a fuzzy method of the data preliminary classification aimed at selection of the data important for the planning and control decisions while meaningless data or data that do not contain new information are discarded. As a result the amount of the resource intensive intellectual processing and the workload in the network infrastructure could be reduced significantly. The paper presents the analysis for the preliminary classification problem, the abstract model of a sensor and classification method. The method is based on the delay which is controlled by multiplicative decrease additive increase rule and expresses the expectation of the context uncertainty. The delay decreases if the variables measured by the sensor presumably change and increases if it keep its value. Numerical examples are presented as well. They consider the set of mount chest accelerometer uncalibrated measurements. We compare source data set and set of the selected values formed by an algorithm implementing fuzzy preliminary classification. Several metrics were used to compare the selected values with the source data and with CMA filter as well.

Application Research on the Use of Multiple Comparison Methods in The Evaluation of the Effectiveness of Total Nitrogen and Total Phosphorus Reduction Measures

With the problem of water pollution becoming increasingly serious, excessive discharge of nitrogen and phosphorus has become a key factor affecting water quality and ecological balance. In order to evaluate the effectiveness of different pollution reduction measures in total nitrogen and total phosphorus control, this study evaluates multiple best management practices (BMPs) using multiple comparison methods, in particular Tukey HSD testing. The results of the statistical analysis of the reduction effects of the different measures showed that all the treatment groups with measures were significantly better than the control group without measures. In particular, BMP 8, BMP 7 and BMP 6 showed the best reduction effects. This study not only provides a scientific basis for pollution control decisions, but also lays the foundation for follow-up research, highlighting the importance of multiple comparison methods in environmental science.

Communicated priors tune the perception of control

Action allows us to shape the world around us. But to act effectively we need to accurately sense what we can and cannot control. Classic theories across cognitive science suppose that this ‘sense of agency’ is constructed from the sensorimotor signals we experience as we interact with our surroundings. But these sensorimotor signals are inherently ambiguous, and can provide us with a distorted picture of what we can and cannot influence. Here we investigate one way that agents like us might overcome the inherent ambiguity of these signals: by combining noisy sensorimotor evidence with prior beliefs about control acquired through explicit communication with others. Using novel tools to measure and model control decisions, we find that explicit beliefs about the controllability of the environment alter both the sensitivity and bias of agentic choices; meaning that we are both better at detecting and more biased to feel control when we are told to expect it. These seemingly paradoxical effects on agentic choices can be captured by a computational model where expecting to be in control exaggerates the sensitivity or ‘gain’ of the mechanisms we use to detect our influence over our surroundings – making us increasingly sensitised to both true and illusory signs of agency. In combination, these results reveal a cognitive and computational mechanism that allows public communication about what we can and cannot influence to reshape our private sense of control.

Effect of evidence-based nursing practices on individualised care: A cross-sectional descriptive study.

This study was conducted to determine the effect of nurses' attitudes towards evidence-based practices on individualised nursing care. This study is a cross-sectional descriptive trial. The descriptive analysis included 200 clinical nurses working in a private hospital between April and September 2022. Data were collected with a personal information form, Individualised Care Scale (A-Nurse Version), and Evidence-Based Nursing Attitude Questionnaire. The relationship between the scales and effect was examined by Pearson correlation and linear regression analyses. T-test, one-way analysis of variance (ANOVA), and post hoc (Tukey, LSD) analysis were used in the statistical analysis of nurses according to their descriptive characteristics. This study has adhered to the STROBE checklist for reporting. They had high mean scores on Individualised Care Scale total (3.68 ± 1.25) and from Clinical Situation (3.78 ± 1.30) and Decisional Control (3.82 ± 1.35) subscales and average score from the Personal Life (3.32 ± 1.29) subscale. Their mean score from the Evidence-Based Nursing Attitude Questionnaire was average (47.64 ± 10.99). There was a positive moderate (r = 0.50, p = 0.000 < 0.05) significant correlation between the scales. Positive attitude towards evidence-based nursing practices positively affects individualised care. Variables such as professional experience positively affect nurses' attitudes towards evidence-based nursing. A positive and significant relationship was found between nurses' attitudes towards evidence-based practice and their attitudes towards individualised care.

Decentralized control of ideal ternary reactive distillation column with inert

Decentralized control of an ideal hypothetical ternary reactive distillation column with an inert component is explored. Both composition measurement based and temperature inferential control structures are designed using simple heuristic approaches (two and three-point). The three-point composition control structure is proposed for the example RD column for the first time in this work. Although stable closed loop responses are seen for the throughput changes for all, the two-point structures have failed to achieve tight control of the product purity in the bottoms or purity of inert in the distillate for the inert composition changes due to fixed reflux ratio. The performances of the three-point control structures for the inert composition changes are quite satisfactory due to the indirect manipulation of the reflux ratio. The independent manipulation of the reflux rate and distillate (instead of fixed reflux ratio policy) is an important control decision for the successful regulation of the example RD column.

Developing and testing personalised nutrition feedback for more sustainable healthy diets: the MyPlanetDiet randomised controlled trial protocol.

Agriculture and food production contribute to climate change. There is mounting pressure to transition to diets with less environmental impact while maintaining nutritional adequacy. MyPlanetDiet aimed to reduce diet-related greenhouse gas emissions (GHGE) in a safe, nutritionally adequate, and acceptable manner. This paper describes the trial protocol, development, and testing of personalised nutrition feedback in the MyPlanetDiet randomised controlled trial (RCT). MyPlanetDiet was a 12-week RCT that provided standardised personalised nutrition feedback to participants based on new sustainable healthy eating guidelines (intervention) or existing healthy eating guidelines (control) using decision trees and corresponding feedback messages. To test the personalised nutrition feedback, we modelled a sample of 20 of the MyPlanetDiet participants baseline diets. Diets were modelled to adhere to control and intervention decision trees and feedback messages. Modelled nutrient intakes and environmental metrics were compared using repeated measure one-way analysis of covariance. Intervention diets had significantly lower (p < 0.001) diet-related GHGE per 2500 kilocalories (kcal) (4.7kg CO2-eq) relative to control (6.6kg CO2-eq) and baseline (7.1kg CO2-eq). Modelled control and intervention diets had higher mean daily intakes of macronutrients (carbohydrates, fibre, and protein) and micronutrients (calcium, iron, zinc, and iodine). Modelled control and intervention diets had lower percent energy from fat and saturated fat relative to baseline. Adherence to the MyPlanetDiet personalised nutrition feedback would be expected to lead to better nutrient intakes and reduced diet-related GHGE. The MyPlanetDiet RCT will test the effectiveness and safety of personalised feedback for a more sustainable diet. Clinical trials registration number: NCT05253547, 23 February 2022.

Robust Model Free Adaptive Predictive Control for Wastewater Treatment Process With Packet Dropouts.

External disturbances and packet dropouts will lead to poor control performance for the wastewater treatment process (WWTP). To address this issue, a robust model-free adaptive predictive control (RMFAPC) strategy with a packet dropout compensation mechanism (PDCM) is proposed for WWTP. First, a dynamic linearization approach (DLA), relying only on perturbed process data, is employed to approximate the system dynamics. Second, a predictive control strategy is introduced to avoid a short-sighted control decision, and an extended state observer (ESO) is used to attenuate the disturbance effectively. Furthermore, a PDCM strategy is designed to handle the packet dropout problem, and the stability of RMFAPC is rigorously analyzed. Finally, the correctness and effectiveness of RMFAPC are verified through extensive simulations. The simulation results indicate that RMFAPC can significantly reduce IAE by 0.0223 and 0.1976 in two scenarios, regardless of whether the expected value remains constant or varies. This comparison to MFAPC demonstrates the superior robustness of RMFAPC against disturbances. The ablation experiment on PDCM further confirms its capability in handling the packet dropout problem.

Theoretical basics of the self-learning system of intelligent locomotive decision support systems

Analysis of works in the field of artificial intelligence allows to make an assumption that today there is a sufficiently developed theoretical basis for the development of intelligent control systems for locomotive control. This will minimize the risks associated with the human factor on the railways. The paper presents the theoretical rationale for the development of a knowledge base for intelligent locomotive control systems. The approach and structure of the self-learning system of intelligent DSS is proposed, the advantage of which is the presence of a fuzzy classifier that works according to the set criteria and determines a fuzzy image of the current train situation. Learning a fuzzy classifier consists in finding a vector K that minimizes the distance between the results of logical inference and experimental data from the sample. The knowledge base is implemented using linguistic variables formalized by methods of fuzzy logic. The use of linguistic values makes it possible to design the base using the usual language of communication, which greatly simplifies both the design process itself and the analysis of the system's performance. Also, the knowledge base has the possibility of constant self-improvement. This happens in two ways. The first is by adding new rules to the knowledge base in case the current situation does not match the existing ones in the base, in which case an additional rule is created and checked for adequacy. The second way is a mechanism for ranking rules in the knowledge base. If the control action of the locomotive driver coincided with the recommendation of DSS in the current situation, then the rating of this recommendation (rule) increases, and in the future the rule selection algorithm will choose one or another control action for the current situation that has the highest rating (that is, it has already been verified several times person). The experiment has shown that the use of intelligent DSS has positive results. On average, the DSS made the correct train control decisions faster than the locomotive driver.

Assessing vaccine strategies for mpox outbreak in New York City using an age-structure model

BackgroundSince May 7 2022, mpox has been endemic in many countries which has attracted the attention of health authorities in various countries and made control decisions, in which vaccination is the mainstream strategy. However, the shortage of vaccine doses and the reduction of protective efficacy have led to unresolved issues such as vaccine allocation decisions and evaluation of transmission scale.MethodsWe developed an epidemiological model to describe the prevalence of the mpox virus in New York City and calibrated the model to match surveillance data from May 19 to November 3, 2022. Finally, we adjusted the model to simulate and compare several scenarios of non-vaccination and pre-pandemic vaccination.ResultsRelative to the status quo, if vaccination is not carried out, the number of new infections increases to about 385%, and the transmission time will be extended to about 350%, while if vaccinated before the epidemic, the number of new infections decreases to 94.2-96%.ConclusionsThe mpox outbreak in New York City may be linked to the Pride event. However, with current vaccine coverage, there will be no more large-scale outbreaks of mpox, even if there is another similar activity. For areas with limited vaccines, priority is given to high-risk groups in the age group [34–45] years as soon as possible.

Intelligent beehive monitoring system based on internet of things and colony state analysis

Bees play a crucial role in terrestrial ecosystems. However, beekeepers are unable to monitor the state of beehives (bees and environment) all the time, which often results in bees escaping or even dying. Currently, some researchers provided the scheme of intelligent beehive monitoring system equipped with the Internet of Things (IoT), There remain two challenges: accurately monitor the environmental status around the hive and accurately track and monitor bees in real time. With the development of the IoT and computer vision algorithms, we hope to provide an automated and efficient system to meet the above challenges. In this paper, we proposed a hive monitoring system, and build a visualization module in the cloud to monitor the activity of bee colonies and the environmental dynamic changes. (1) We proposed a multi-bee tracking algorithm to solve the problem of monitoring bees at the door of the hive; (2) we constructed a dataset containing various complex scenes, named BEE22, for training and testing the performance of our algorithm; (3) we designed a bee counting rule, based on results of multi-bee tracking algorithm, to reasonably count the bees entering or leaving the beehive; (4) we have deployed multiple sensors around(center, margin, door, and environment) the hive to accurately reflect the changes in the environment around the hive. Experimental results demonstrate the effectiveness and excellence of our system. In particular, the tracking performance of the multi-bee tracking algorithm reaches 83.5 % ± 0.7 % Multiple Object Tracking Accuracy (MOTA) and 77.3 %±0.2 % Multiple Object Tracking Precision (MOTP), speeds up to 16 frames per second, compared with other algorithms, MOTA and Identity F1 Score (IDF1) are improved by 5.4 % and 8.2 % respectively. Moreover, our counting algorithm also achieved excellent results, with root mean square error (RMSE) of 1.3 ± 0.1, 0.2 ± 0.0, and 1.6 ± 0.1 in counting the number of bees current, entry, and out scene in an episode, respectively. After that, the system will be deployed and monitored for a long time in the actual scenario, it was found that the activity of bees decreased significantly under heavy rainfall conditions. Additionally, the activity of the bee colony will also increase accordingly, when the amplitude is 500 dB to 2000 dB, the temperature of the center of the beehive is 25 °C to 37 °C, and the humidity is 48 % to 67 %. In summary, our system can provide valuable information for bee farmers to make control decisions on hives.

Quantifying the Influence of Climatic and Anthropogenic Factors on Multi-Scalar Streamflow Variation of Jialing River, China

Clarifying the impact of driving forces on multi-temporal-scale (annual, quarterly and monthly) runoff changes is of great significance for watershed water resource planning. Based on monthly runoff data and meteorological data of the Jialing River (JLR) during 1982–2020, the Mann–Kendall tendency testing approach was first applied to analyze variation tendencies of multi-timescale runoff. Then, abrupt variation years of runoff were determined using Pettitt and cumulative anomaly mutation testing approaches. The ABCD model was employed for simulating hydrological change processes in the base period and variation period. Finally, influences of climatic and anthropic factors on multi-scalar runoff were computed using the multi-scalar Budyko formula. The following conclusions were drawn in this study: (1) The mutation year of discharge was 1993; (2) the monthly runoff in the JLR presented a “single peak” distribution, and the concentration degree and concentration period in the JLR both showed an insignificant reduction trend; (3) anthropic factors were the dominant factor for spring runoff variations; climatic factors were the dominant factor on annual, summer, fall and winter runoff variations; (4) except for November, climatic factors were the dominant factor causing runoff changes in the other 11 months. This study has important reference value for water resource allocation and flood control decisions in the JLR.

The Role of Nudges and Behavioral Interventions in Financial Decision Making

Financial decision-making, both at individual and institutional levels, plays a critical role in determining economic outcomes. Traditional economic theory posits that individuals are rational agents who make decisions based on complete information and a consistent set of preferences. However, behavioral economics has challenged this notion, highlighting cognitive biases, heuristics, and social influences that can drive suboptimal decisions. Nudges and behavioral interventions have emerged as tools to guide individuals toward better financial choices without restricting their freedom of choice. The purpose of this research is to present a descriptive investigation of behavioral interventions and nudges in financial decision-making. The main cognitive biases influencing financial behavior are first described, and then several nudge kinds and their applicability in diverse financial contexts— retirement savings, debt management, spending control, and investment decisions—are explored. Along with reviewing empirical data on nudges' efficacy, the paper addresses the ethical issues surrounding their application and pinpoints the variables that affect their effectiveness. This study aims to educate consumers, financial institutions, and policymakers about the potential advantages and constraints of behavioral interventions by providing a thorough overview of how nudges might impact financial decision-making. Keywords: Nudges, Behavioral Intervention, Financial Decision-Making, Cognitive Biases, Financial Literacy.

Occurrence and fate of five representative neonicotinoid insecticides across different wastewater treatment plants and the impact on receiving water bodies

Neonicotinoids (NEOs), despite their widespread use as insecticides, exhibit a notable knowledge deficit in regards to their presence in wastewater treatment plants (WWTPs) and their surrounding environments. This study delves into the presence and disposition of 5 NEOs: Thiamethoxam (THM), Clothianidin (CLO), Imidacloprid (IMD), Acetamiprid (ACE), and Thiacloprid (THA) across 3 domestic WWTPs and their receiving waters. Notably, THM, CLO, and ACE were consistently detected in all water and sludge samples, with THM emerging as the most abundant compound in both influent and effluent. Among the 3 WWTPs, WWTP 2, employing a fine bubble oxidation process, achieved the highest removal efficiency, surpassing 68%, in contrast to WWTP 1 (CAST) at 37% and WWTP 3 (A/A/O) at 7%. Biodegradation played a pivotal role in NEO removal, accounting for 36.7% and 68.2% of the total removal in WWTP 1 and WWTP 2, respectively. Surprisingly, in WWTP 3, biotransformation process inadvertently increased ACE and CLO concentrations by approximately 4.1% and 4.5%, respectively. The total NEO concentration in the receiving surface waters ranged from 72.7 to 155.5 ng/L, while sediment concentrations were significantly lower, spanning between 0.10 and 1.53 ng/g. WWTPs serve as both a removal and concentration point for NEOs, thereby significantly influencing their transportation. Additionally, the concentration of most NEOs in the receiving waters progressively increased from upstream to downstream, highlighting the substantial impact of WWTP discharges on natural water environments. This research offers valuable insights into NEO pollution surrounding WWTPs in the Pearl River Delta, ultimately aiding in pollution control and environmental protection decisions.

Sensing with shallow recurrent decoder networks

Sensing is a universal task in science and engineering. Downstream tasks from sensing include inferring full-state estimates of a system (system identification), control decisions and forecasting. We propose a shallow recurrent decoder (SHRED) neural network structure for sensing, which incorporates (i) a recurrent neural network to learn a latent representation of the temporal dynamics of the sensors and (ii) a shallow decoder that learns a mapping between this latent representation and the high-dimensional state space. SHRED enables accurate reconstructions with far fewer sensors, outperforms existing techniques when more measurements are available and is more robust to random sensor placements. In the example cases explored, complex spatio-temporal dynamics are characterized by exceedingly limited sensors that can be randomly placed with minimal loss of performance.

TrafPS: A shapley-based visual analytics approach to interpret traffic

Recent achievements in deep learning (DL) have demonstrated its potential in predicting traffic flows. Such predictions are beneficial for understanding the situation and making traffic control decisions. However, most state-of-the-art DL models are considered “black boxes” with little to no transparency of the underlying mechanisms for end users. Some previous studies attempted to “open the black box” and increase the interpretability of generated predictions. However, handling complex models on large-scale spatiotemporal data and discovering salient spatial and temporal patterns that significantly influence traffic flow remain challenging. To overcome these challenges, we present TrafPS, a visual analytics approach for interpreting traffic prediction outcomes to support decision-making in traffic management and urban planning. The measurements region SHAP and trajectory SHAP are proposed to quantify the impact of flow patterns on urban traffic at different levels. Based on the task requirements from domain experts, we employed an interactive visual interface for the multi-aspect exploration and analysis of significant flow patterns. Two real-world case studies demonstrate the effectiveness of TrafPS in identifying key routes and providing decision-making support for urban planning.

Game-Based Flexible Merging Decision Method for Mixed Traffic of Connected Autonomous Vehicles and Manual Driving Vehicles on Urban Freeways

Connected Autonomous Vehicles (CAVs) have the potential to revolutionize traffic systems by autonomously handling complex maneuvers such as freeway ramp merging. However, the unpredictability of manual-driven vehicles (MDVs) poses a significant challenge. This study introduces a novel decision-making approach that incorporates the uncertainty of MDVs’ driving styles, aiming to enhance merging efficiency and safety. By framing the CAV-MDV interaction as an incomplete information static game, we categorize MDVs’ behaviors using a Gaussian Mixture Model–Support Vector Machine (GMM-SVM) method. The identified driving styles are then integrated into the flexible merging decision process, leveraging the concept of pure-strategy Nash equilibrium to determine optimal merging points and timing. A deep reinforcement learning algorithm is employed to refine CAVs’ control decisions, ensuring efficient right-of-way acquisition. Simulations at both micro and macro levels validate the method’s effectiveness, demonstrating improved merging success rates and overall traffic efficiency without compromising safety. The research contributes to the field by offering a sophisticated merging strategy that respects real-world driving behavior complexity, with potential for practical applications in urban traffic scenarios.