Please enable JavaScript.

Coggle requires JavaScript to display documents.

Opal Model (Regression (Route-to-route change based (Each route contains…

- - - - Each link has its properties.
        And these properties formulate the travel cost for each route
        
        Flow on link
        
        Capacity of link
        
        Free flow time
        
        Cost function integrates with actual flow swapping rate, giving the model
- - - - Perception of Travel Time
        with respect to each Time Range
        
        The travel information observed from the individual's connected social network assists
        travellers to grasp a better perception of travel time w.r.t each time range
        
        The day-to day updates allows the Bayesian updating model to be manipulated.
        
        Finally, the utility function (f(travel time)) determines the performance of each time range. And the Logit model calculates the probability of a certain time range which could be utilised by the traveller i based on his or her updated perception on a certain day t.
        
        Simultaneously, their own travel experiences (on t=t-1) also helps them to judge the travel time
- - - - What we need to comply with the existing models?
        
        What is not available
        
        Preferred destination arrival time
        
        Early and late penalties
        
        What we could get?
        
        A revised utility function complying with the nature of public transport system.
        
        Directions
        
        Factors affect travellers' choice of departure time range?
        
        Expected time of arrival* Cannot be determined
        
        In vehicle time
        
        perceived in-veh time
        
        Distribution of perceived
        
        Delays
        
        perceived delay
        
        experiences and previous day condition
        
        this may requires the distribution of the delay w.r.t. a specific time range -> time range
        
        Service headway
        
        we can assume this is fixed and travellers know the scheduled time well.
        
        Number of departure travellers at each time range is known.
        
        Could we use the probability of choosing a certain range to determine the cost function?
    - - Delay
        
        On
        
        Off
      - In vehicle Time (travel time) -> this counts the traffic conditions, which varies w.r.t. days and time ranges