In order to reduce the waiting time of buses at signal-controlled intersections and enhance bus operating efficiency, several scholars have conducted research on transit priority control, starting with intersection signal priority. Tarikul developed a dynamic priority control system that balances signal control demands and transit priority, aiming to reduce the average person delay at intersections [5]. Qiao proposed an optimization model for transit signal priority at a signalized intersection based on the phase clearance reliability index, with the objective of minimizing average person delay [6]. Truong presented an advanced transit signal priority (ATSP) control model that considers the arrival distributions of buses at downstream intersections when providing priority at upstream intersections, resulting in an approximate 10% improvement in bus line efficiency [7]. Li reduced the average person delay and improved the traffic efficiency of the trunk line by implementing a signal priority control method based on coordinating green waves [8].

Scholars have also studied transit priority from the perspective of pre-signal settings, considering that it may reduce the operating efficiency of social vehicles. He aimed to reduce the average person’s delay by prioritizing bus signals through adaptive control and pre-signal settings [9]. Liang analyzed the queuing situations at bus intersections using the distributed wave theory and designed a pre-signal control algorithm for transit priority based on queuing length to achieve non-stop traffic at intersections [10]. Bie developed a coordinated control algorithm between the main signal and pre-signal of bus intersections based on the pre-signal of transit priority, thereby reducing the impact of transit priority phases on social vehicle efficiency [23]. While these studies reduced bus delays at intersections to some extent, they failed to improve the reliability of bus punctuality, making it difficult to further enhance the punctuality service level of buses.

With the advancement of vehicle–road cooperative technology, real-time speed guidance for bus vehicles has become possible. This development has prompted scholars to explore improvements in bus service levels through the lens of speed guidance. Khaled achieved further reductions in bus delays by implementing a strategy of early braking at red lights and extending green lights based on speed guidance while incorporating transit signal priority [24]. Shu solved direct bus and left turn bus priority control using speed guidance under the premise of transit signal priority, improving bus traffic efficiency at intersections [11]. Chiara determined different speed guidance strategies by analyzing the priority of bus formation and independent buses, reducing average person delay [12]. Deng proposed a dynamic real-time speed guidance model to mitigate operation delays of bus lines caused by signalized intersections and uneven road conditions [13].

Scholars also studied speed guidance from the perspective of improving bus punctuality since it enhances the bus operation reliability. Takashi developed an inter-station control strategy that coordinates speed guidance and signal control based on punctuality demands for bus arrivals, leading to improved reliability of bus arrivals [14]. Yan proposed a real-time bus speed control strategy to minimize the mean absolute error of bus headway due to unstable bus arrival times, significantly enhancing the punctuality reliability of bus arrivals [15]. Zhang analyzed bus trajectory data and implemented guidelines to achieve a balanced distribution of bus headway, thereby improving bus operation reliability [16]. While these studies have made improvements in enhancing the reliability of bus operations, optimizing speed guidance solely based on a single road section may lead to local optimization but poor overall conditions.

The key to improving bus punctuality and operational efficiency involves setting the bus schedule, prompting scholars to focus on its optimization. Li developed a public transport scheduling model for a microsystem that aims to minimize passenger waiting time while maximizing the number of passengers per bus. This is achieved by optimizing departure intervals and utilizing both traditional and rapid buses simultaneously [17]. Gkiotsalitis constructed an optimization model for bus travel schedules based on passenger demand and travel time expectations [18]. Shang proposed an evaluation model that takes into account both passenger satisfaction and traffic efficiency. This model was used to optimize and adjust the bus operating schedule, and its effectiveness was demonstrated through a case study conducted in Beijing [25]. Banerjee used a school bus as an example and proposed a scheduling model aimed at optimizing the operational efficiency of bus schedules [19]. Teng established a multi-objective optimization model to balance departure headway, reduce the number of vehicles used, and lower electric bills. This model optimized bus operation schedules under multiple constraints, including limits on the range of departure intervals, the number of available vehicles, and bus endurance mileage at different periods [20]. To optimize the bus schedule, Liu constructed a super-efficient DEA model based on indicators of passenger waiting time and congestion [21]. Ma proposed a dynamic schedule optimization scheme based on the correlation of passenger time demand and travel time between stations, utilizing bus GPS data and IC card data [26]. Zhang developed an optimal design model to minimize passenger transfer waiting time, considering the importance of transfer stations and different travel time utility values for passengers with different travel purposes, optimizing the bus schedule using genetic algorithms [22]. While these studies improved bus operation efficiency and passenger satisfaction to some extent, they rarely optimized and adjusted bus capacity allocation and schedule schemes from the perspective of passenger travel demand, making it challenging to optimize passenger platform retention.

Existing research has made progress in optimizing bus service levels at various levels, yielding some favorable outcomes. However, most punctuality controls focus on optimizing a single section, which overlooks the optimization of bus punctuality across the entire line. As a result, there is a risk of local optimization with overall poor performance, affecting bus punctuality service levels. Furthermore, existing research often neglects the correlation between passenger travel demand and bus capacity allocation, which can lead to a mismatch between supply and demand, thereby affecting the quality of bus service.