3.4 Conflict Detection Module
3.4.1 Conflict Detention Mechanism
Adjusting the departure time of a train may cause a chain reaction to other trains. An adjusted train that is affected by the resolution of a certain conflict may influence the following trains and lead to more conflicts. Thus, an inserted train may bring about multiple conflicts related to different trains. Deciding the resolution order of conflicts is important. Thus, this study developed the CDM process to be an event ordering standard and a concept of a conflict box to solve the chain reaction problem. The CDM has three main parts as presented in Figure 3.10.
The CDM procedure begins after implementing the TIM. The above flowchart process has the following settings, “I” which is the code of the conflict-causing train, “j”
which is the code of the conflict-affected train, “k” is the number of conflict boxes this study developed to resolve the chain reaction situation based on the meaning of the one-resolution loop, and “TCk” which is the code of the conflict-causing train in solving the conflict box k situation. When adding a new train to the analyzed diagram, Part 1 is first executed. This part mainly focuses on the detection of the conflict-causing train i and on whether or not it causes conflict to a conflict-affected train j. If a conflict-causing train causes conflict, all of the conflicts are placed inside the Conflict Box k to memorize the events. The first conflict is resolved following the FCFS rule. Once conflict occurs, Part 2 addresses the first conflict using the CRM and then develops the concept of recursion by opening a new Conflict Box k+1, which represents the chain reaction conflict. This type of Conflict Box k+1, where k is not equal to 1, stands for the train chain reaction
conflict-causing train does not cause any conflict or if all of the conflict events are resolved, the Conflict Box k is closed until the end of the detention process of the influence of TCk, and the CDM process proceeds to Part 3. If the code of closed Conflict Box k does not equal to 1, it indicates that some unsolved conflict remains in other conflict boxes. Thus, the process must return to the previous Conflict Box k-1 to discuss the surplus conflict. The pseudo code of the CDM is proposed below.
Pseudo code of the Conflict Detention Module (CDM) 1: Input: i= Code of input train, k = code of conflict box,
2:
TC
k= code of conflict-causing train in solving Conflict Box k 3: k = 0,TC
k= i4: Function Conflict_Happened (conflict-causing train i) 5: k = k+1
6: For each train j existing in diagram 7: If train i conflict train j then
8: Deal with Conflict Resolution Module 9: Conflict_Happened ( j )
10: End If
11: Next
12: End For 13: End Function
3.4.2 Example of a Conflict Detection Module Working Process
The concepts of CDM and CRM are complicated to understand using a single description. The following content provides an example of the procedure of the CDM and the simulation process. Figure 3.11 shows the original conflict-resolved result that has three southbound trains operating in a single-track corridor with four stations. Each blocking time stairway represents a movement behavior of a certain train. The absence of
an overlapping condition in the diagram means that no conflict has happened in this operation.
Figure 3.11 Schematic Diagram of the Conflict Detection Module Example (1) An opposite direction train that operates northbound is added to the existing diagram, as shown in Figure 3.12, in a blue blocking time stairway. Following the TIM, the top of the first blocking time diagram of Train 4 is set to zero.
Figure 3.12 Schematic Diagram of the Conflict Detection Module Example (2) Two conflicts occur as Train 4 is placed into the existing diagram as shown in Figure 3.13. The two conflicts caused by Train 4 are Conflict 1 of the first conflict-affected train that occur at Block 3 and Conflict 2 of the third conflict-affected train that occur at Block 5.
Figure 3.13 Schematic Diagram of the Conflict Detection Module Example (3) Given that it is the first detection procedure for adding a train, Conflict Box 1 has been opened to memorize the above conflict events caused by Train 4. The order of the occurred multiple conflict events is decided on the basis of the time each conflict happened, and each conflict is loaded into Conflict Box 1 after the sequence as shown in Figure 3.14. By following the FCFS rule, the CDM decides to put the first conflict into the CRM for resolution. According to the basic operation rule for crossing situations, the departure time of Train 4 at Station C (td,4) is earlier than that of Train 1 at Station B(td,1).
Therefore, the departure time of Train 1 at Station B shifts backward to after the arrival time of Train 4 at Station B. Before the time movement of Train 1, the CRM should check and keep the station track suitable for train crossing and the single and total waiting time within the limitation. Once the conflict event is resolved, the CDM removes this event from Conflict Box 1 and opens another Conflict Box 2 to record the following chain reaction conflict caused by the time shift in Train 1. Figure 3.14 presents the adjusted diagram and conflict boxes.
Figure 3.14 Schematic Diagram of the Conflict Detection Module Example (4) Two conflicts happen after the CDM process because the time shift of Train 1 affects Train 2, which is located at Blocks 4 and 5. The same as the previous procedure, the CDM places the conflicts into Conflict Box 2 and solves them using the CRM. The result is shown in Figure 3.15.
Figure 3.15 Schematic Diagram of the Conflict Detection Module Example (5) The conflict is resolved by moving the arrival time of Train 2 at Block 4 (ta,2) backward to the departure time of Train 1 at Block 4 (td,1) to avoid the overlapping condition at Block 4 to Train 1. This measure also cleans Conflict 4, removes the conflict events from Conflict Box 2, and creates a new Conflict Box 3 to detect the following chain reaction conflict caused by the time shift of Train 2. Conflict Box 2 is empty without any conflict events. A conflict box that becomes empty is closed by the CDM to show that no conflict exists given the time shift influence of the previous train. Figure 3.16 shows the diagram, the conflict boxes of the CDM, and the simulation process.
Figure 3.16 Schematic Diagram of the Conflict Detection Module Example (6) Similar to the previous analysis process, the CDM detects Conflicts 5 and 6 of Train 3 located at Blocks 4 and 5, which are affected by the shift of Train 2 and the conflicts placed into Conflict Box 3. This conflict can be solved by moving the arrival time of Train 3 at Block 4 (ta,3) backward to the departure time of Train 2 at Block 5 (td,2). The result is shown in Figure 3.17.
Figure 3.17 Schematic Diagram of the Conflict Detection Module Example (7) Given that Conflict 5 has been resolved, this measure also clears Conflict 6, removes these conflicts from Conflict Box 3, and creates Conflict Box 4 to detect the following chain reaction conflict caused by the time shift of Train 3. Conflict Box 3 becomes empty, and the CDM closes it. Figure 3.18 shows the current diagram and conflict boxes.
Figure 3.18 Schematic Diagram of the Conflict Detection Module Example (8) The CDM also detects whether or not the time shift event of Train 3 causes any of the following conflicts. Given that the process shows no conflict, Conflict Box 4 is closed, and the CDM procedure returns to Conflict Box 1 to resolve the remaining conflict.
Therefore, Conflict 2 is taken into the CRM, and the departure time of Train 4 at Station B (td,4) is later than the departure time of Train 3 at Station A (td,3), which follows the basic operation rule for the crossing situation. Therefore, the departure time of Train 4 at Station B shifts backward to after the arrival time of Train 3 at Station B. The result is presented in Figure 3.19.
Figure 3.19 Schematic Diagram of the Conflict Detection Module Example (9) After Conflict 2 has been resolved, all of the conflict boxes become empty and do not cause other conflict boxes. Thus, no conflict will occur, and the CDM procedure is completed. The final result is shown in Figure 3.20, and the details of the total decision process and result by the CDM procedure are presented in Table 3.1.
Figure 3.20 Schematic Diagram of the Conflict Detection Module Example (10) Table 3.1 Conflict Resolution Order and Result of the CDM Procedure
3.5 ChapterSummary
This study develops an easy-to-use simulation method to analyze the route capacity mixed with single and double tracks. The standard operation times of each section, operation margin, infrastructure condition, and several basic train operation rules are the inputs and references used to conduct the simulation process. This simulation method contains the concept of blocking time diagram and blocking time stairway to detect the conflict using a diagram of the overlapped situation to avoid the deadlock problem
efficiently. Three main modules are established in the simulation method: TIM, CDM, and CRM. The TIM focuses on inserting a new blocking time stairway to an existing diagram. The CDM provides such functions as detecting whether or not the current train blocking time stairways causes any train conflict and the judging the order of train conflicts. The CRM is used to solve train conflicts in the simulation process. Given the complicated effect caused by the train chain reaction condition, the conflict box concept is developed to memorize all of the different conflicts caused by different conflict-causing trains in different conflict boxes to help resolve conflict events in a reasonable sequence.
This study provides a small case as an example to help understand the CDM procedure and the simulation process.
61
CASE STUDY
Chapter 3 introduced the concept and simulation process and provided an example to show how the module works in the process. This chapter applies the simulation capacity model to an actual case to estimate the capacity and compare it with the current utilization rate using Microsoft C# with several assumptions. The main operator of the conventional railway of Taiwan, the Taiwan Railway Administration (TRA), is selected as the railway system for the study. Two of the single- and double-track mixed corridors are chosen for the case study to estimate the route capacity. Section 4.1 lists the programing assumptions of the proposed simulation model. Section 4.2 applies the TRA southern corridor from Fangliao to Taitung section to Cases I and II, and Section 4.3 applies the TRA eastern corridor from Hualien to Taitung. Section 4.3 discusses the capacity after a double-track construction project from Hualien to Taitung. Section 4.4 validates the model using the existing continuous single-track capacity model, CRCS.
Section 4.5 presents the sensitivity analysis and discussion. Section 4.6 summarizes the chapter.
4.1 Programming Assumptions of the Simulation Model
This study establishes a new analysis concept of capacity. The programming of this simulation model needs to be developed. However, the actual operation has many elements, and all of the train operation conditions are not considered to avoid complicated programming. Therefore, several assumptions about the simulation model are made as follows.
I. The signal system using a three-aspect signal keeps all trains operating with a full-speed signal.
II. All trains can only depart from two terminal stations of the analyzed corridor.
III. The standard operating time of two adjacent signals comes from the standard operating time of two adjacent stations multiplied by the percentage of the signal distance of the station distance.
AB
TAB = Station standard operation time from Station A to Station B
i
T
AB, = Signal standard operating time of block i between Stations A and BD
AB = Distance from Station A to Station BD
i = Distance of block iB
AB = Set of signal block i between Stations A and BIV. The capacity influence of different station track layouts that cause different departure and arrival headways is not considered.
V. The limitation of station tracks usage is not considered because of the station track layout.
VI. The interlocking sections, which include crossover and level crossing, within the station section are combined as shown in Figure 4.1 and Figure 4.2 in this model.
Figure 4.1 Actual Signal Section with Interlocking and Station Sections
Figure 4.2 Assumed Signal section in the Proposed Simulation Model 4.2 TRA South-Link Line Capacity Analysis
The southern corridor of TRA, which is also called the South-link Line, connects the western and eastern main line of Taiwan between Fangliao and Taitung stations. The South-link Line is 98.2 km long and has 12 stations that provide a train passing function.
It can be divided into 11 sections in both directions as shown in Figure 4.3. This corridor mainly provides the connection between the western and eastern areas of Taiwan rather than caters to the local demand within the area. The percentage of inter-city trains is higher than that of the local trains to supply longer distance trips through transportation
capacity. This corridor was completed in 1991. All sections operate on single-track sections except for the Central Signal Station to Guzhuang Station, which operates on a double-track section for bidirectional trains. Moreover, low population causes low demand and geographical characteristics. The route goes through a series of mountain ranges that leads to long distances between stations, thus making the capacity quite low.
Figure 4.3 Network of the TRA South-Link Line
The following section analyzes the current single- and double-track mixed corridor capacity and compares the conditions of all sections operating on a single track, as well as the section from the Central Signal Station to the Guzhuang Station, to discuss the influence of the sectional track duplication from a single track to a double track. The trains operating in this corridor has two main types. This study sets the composition of inter-city trains to local trains to 4–1 and the operating margin to 0.2 as the basic input of the simulation analysis model because of the low demand of local service.
Scenario 1: All sections are in a single-track condition
Every section in this corridor is assumed to operate in a single-track condition.
Therefore, the result shows that the capacity of the section between the Central Signal Station and Guzhuang Station is not equipped with a double main track. The simulated result is shown in Figure 4.4.
Figure 4.4 Continuous Single-track Capacity of the South-link Line
As the number of simulated trains increases, the calculated result of capacity becomes stable and willing to converge to a certain value. The line capacity between Fangliao and Taitung is 2.268 trains/hour in a single-track condition for bi-directional operating trains. This study collects the train delay result of each direction and each train class from the simulated result to conclude the operating performance in this kind of condition. Each train needs to comply with the limitation of the single and total waiting time. The waiting time represents the original dwelling time plus the additional delay time to accommodate another overtaking or crossing train. The delay time result is shown in Table 4.1.
Table 4.1 Delay of Each Train Class of the South-link Line in a Single-track Condition
* The total waiting limit for an intercity train is 720 s and that for a local train is 1,200 s.
* The single waiting limit for an intercity train is 600 s and that for a local train is 960 s.
Table 4.1 shows that an inter-city train needs to wait for another train for an additional 2 min on average. However, a local train needs to wait 20 min. This wait time and thus causes the public to be less attracted to take a local train. Moreover, this study focuses on the maximum single delay and the total delay of a local train, which total to 1,135 s and 1,130 s of waiting, respectively. This delay time and the original dwelling time of 60 s is nearly equal to the limit of the total waiting time. Thus, the whole operation capacity is restricted by the limitation of waiting time.
Figure 4.5 Blocking Time Diagram of the South-link Line in a Single-track Condition Figure 4.5 presents the simulated blocking time diagram result of the South-link Line in a single-track condition, and this study takes a short period of the result to show the details in Figure 4.6. In the single-track condition, all of the blocking time diagrams do not overlap with each other expect for the station section. Given that the main track does not provide train overtaking or crossing function, many waiting behaviors, such as a local train allowing another train to overtake or cross, occur.
67
Figure 4.6 Detailed Blocking Time Diagram of the South-link Line in a Single-track Condition
68
Scenario 2: Current single- and double-track mixed condition
The tracked condition of the double track between the Central Signal Station and Guzhuang Station is considered. Therefore, this route becomes single- and double-track mixed sections with two single-track sections and a double-track section. The simulated result is illustrated in Figure 4.7.
Figure 4.7 Continuous Single- and Double-track Mixed Capacity of the South-link Line
As the number of simulation trains increases, the simulated result of capacity becomes convergent at 2.421 trains/hour. Compared with the simulated result in the single-track condition, the capacity improves by increasing by 0.2 trains/hour. Whereas capacity increases by 0.2 trains/hour, the average delay time decreases by about 20%. The improvement of the single waiting time shows the most significant decrease of about 40%.
The total waiting limitation remains a major restriction, but the operating efficiency of the train improved greatly. The result is shown in Table 4.2.
Table 4.2 Delay of Each Train Class of the South-link Line in a Mixed-track Condition
* Total waiting limit for an inter-city train is 720 s and that for a local train is 1,200 s.
* Single waiting limit for an inter-city train is 600 s and that for a local train is 960 s.
Figure 4.8 presents the simulated blocking time diagram result of the South-link Line in a single- and double-track mixed condition. A short period of the result is taken to show the details in Figure 4.9. In the mixed-track condition, trains operating on a single track should wait in the station for another train to overtake or cross. However, the double-track section can facilitate train crossing in the main track using each track employed for trains of certain direction. The infrastructure condition of the double track is represented by gray lines. Thus, the train-crossing action is allowed in the double track as shown by the gray lines in Figure 4.10. The waiting times of the local train stop at a station also decrease significantly.
Figure 4.8 Blocking Time Diagram of the South-Link Line in the Mixed-track Condition
70
Figure 4.9 Detailed Blocking Time Diagram of the South-link Line in a Mixed-track Condition
71
4.3 TRA Hualien
–
Taitung Line Capacity AnalysisThe Hualien–Taitung Line is one part of the eastern corridor of the TRA between Hualien and Taitung. A total of 21 stations provide the train passing function, and this 151 km-long route can be divided into 20 sections as shown in Figure 4.10. Hualien and Taitung are major tourism areas in Taiwan. An inconvenient traffic condition is caused by geographic factors, tourism, and returning hometown demand, which are beyond the transportation capacity supplied by the operator during vacations or special days. Two
The Hualien–Taitung Line is one part of the eastern corridor of the TRA between Hualien and Taitung. A total of 21 stations provide the train passing function, and this 151 km-long route can be divided into 20 sections as shown in Figure 4.10. Hualien and Taitung are major tourism areas in Taiwan. An inconvenient traffic condition is caused by geographic factors, tourism, and returning hometown demand, which are beyond the transportation capacity supplied by the operator during vacations or special days. Two