Purpose

The analysis of the DMAN Playbook seeks to better understand the departure and arrival delays that could potentially arise when utilising the published matrix during the operationalisation of the DMAN position. This analysis will look into variables that could increase or decrease the runway capacity such as a) Runway Occupancy Time (ROT), b) the use of 60 second departures, and c) differences in wake turbulence categories.

Data

The analysis will look at data collected for a) ROT, the b) % utilisation of 60 second departures, the c) wake turbulence categories of departures and arrivals, as well as d) the capacity of both runways for departures and arrivals. These are tabulated against the Playbook within the DMAN-AMAN Integration project which proposes strategies for mixed mode operations in Changi to reduce both departure and arrival delays. The proposed DMAN matrices are shown below, broken down into strategies for SIPA and non-SIPA:

Establishing Initial Baseline Capacity

Assumptions

The capacity of the departure runway are calculated based on 90 seconds and 60 seconds departure interval between aircraft of the same wake turbulence category, while the arrival runway’s capacity is based on the minimum in-trail of 3nm. It also assumes Changi in a VMC condition and the absence of runway events such as scheduled inspections, FOD retrieval, birdstrikes, and any other operational restrictions.

These assumptions are necessary in establishing the maximum number of departures and arrivals both runways are able to handle and for further calculations to determine potential departure and arrival delays when utilising the proposed DMAN matrices.

The calculated baseline numbers are summarised in the table below:

Arrival Mode 90 seconds Departure Interval (Per hour) 60 seconds Departure Interval (Per hour) Arrivals per hour
No arrivals 40 60 0
4/1 mode 24 40 8
3/1 mode 20 36 10
2/1 mode 14 20 13

Differences in Departure Delay : 60 seconds vs 90 seconds Departure Interval

This section illustrates the differences in the number of departures and arrivals when a) no arrivals are on the departure runway, b) 4/1 arrival mode, c) 3/1 arrival mode, d) 2/1 arrival mode are currently used as strategies to reduce arrival delays.

Changi Traffic Capacity Per Hour: 90 Seconds Departure Interval

With departures on runway 2 only, the graph above describes the maximum number of departures using 90 second departure intervals (of the same wake turbulence category) with increasing number of arrivals on the departure - based on the arrival mode.

We see that the gradient of the graph remains the same between 4/1 and 3/1 mode - indicating no biasness of capacities towards either departures or arrivals. However, when the arrival mode increases to anything greater than 3/1 mode, there is an obvious preference towards arrivals.

A similar graph for departures with 60 seconds departure interval is shown below. It shows that even with increasing arrival modes, a consistent 60 second departure interval does not show any greater preference to either departure or arrivals.

Changi Traffic Capacity Per Hour: 60 Seconds Departure Interval

The difference between the two are clearly shown in the plot below:

Impact of Playbook Matrix Strategies to Changi Departures & Arrivals

Proposed Playbook Matrix: 90 seconds Departure Interval, SIPA Operations

The playbook matrix establishes the strategies necessary to tackle arrival demand with its corresponding departure demand (and vice versa). With these in mind, calculations for the departure capacity were made and the number of departures vs arrival delays, as well as in delays in minutes were obtained.

The two graphs are shown below:

The horizontal and vertical lines indicate the definitions of departure delays (>20 minutes) and arrival delays (>15 minutes) respectively as a visual reference.

Proposed Playbook Matrix: 90 seconds Departure Interval, Non-SIPA Operations

Similar calculations were made for strategies employed during Non-SIPA operations, and they are shown in the figures below:

Possible Changes to Playbook Matrix

Proposal 1: Alternating SIPA/Arrival Mixed Mode Spacing

SIPA Operations
Playbook Matrix Arrival Demand Departure Demand Current Proposed Strategy Change to Proposed Strategy
A1 >=39 > =34 11 nm 11/15 nm (alt)
A2 >=39 32-33 11 nm 11/15 nm (alt)
A3 >=39 30-31 11 nm 11/15 nm (alt)
A4 >=39 28-29 7 nm 7 nm
A5 >=39 27-26 7 nm 7 nm
A6 >=39 <=25 7 nm 7 nm
B1 35-38 > =34 15 nm 15 nm
B2 35-38 32-33 11 nm 11/15 nm (alt)
B3 35-38 30-31 11 nm 11/15 nm (alt)
B4 35-38 28-29 11 nm 11 nm
B5 35-38 27-26 11 nm 11 nm
B6 35-38 <=25 7 nm 7 nm
C1 32-34 > =34 15 nm 15 nm
C2 32-34 32-33 15 nm 15 nm
C3 32-34 30-31 11 nm 11 nm
C4 32-34 28-29 11 nm 11 nm
C5 32-34 27-26 11 nm 11 nm
C6 32-34 <=25 7 nm 7 nm

Proposal 2: Increasing SIPA spacing for the highest departure & arrival mix (Matrix A1)

Playbook Matrix Arrival Demand Departure Demand Current Proposed Strategy Change to Proposed Strategy
A1 >=39 >=34 11 nm 15 nm
A2 >=39 32-33 11 nm 11 nm
A3 >=39 30-31 11 nm 11 nm
A4 >=39 28-29 7 nm 7 nm
A5 >=39 27-26 7 nm 7 nm
A6 >=39 <=25 7 nm 7 nm
B1 35-38 > =34 15 nm 15 nm
B2 35-38 32-33 11 nm 11 nm
B3 35-38 30-31 11 nm 11 nm
B4 35-38 28-29 11 nm 11 nm
B5 35-38 27-26 11 nm 11 nm
B6 35-38 <=25 7 nm 7 nm
C1 32-34 > =34 15 nm 15 nm
C2 32-34 32-33 15 nm 15 nm
C3 32-34 30-31 11 nm 11 nm
C4 32-34 28-29 11 nm 11 nm
C5 32-34 27-26 11 nm 11 nm
C6 32-34 <=25 7 nm 7 nm

When comparing the two:

Proposal 3: Simplified Playbook Matrices

A simplified version of the ATC Playbook Matrix is also proposed, accounting for >25 departures and >25 arrivals per hour.

ATC Playbook Matrix for SIPA Operations
ATC Playbook Matrix for Non-SIPA Operations

Using these matrices we obtain plots similar to the ones above:

In summary:

Matrix Departure Demand Arrival Demand SIPA Strategy
A1* 41-45 41-45 11/15nm (alt)
A2* 36-40 41-45 11/15nm (alt)
A3* 31-35 41-45 11nm
A4* 26-30 41-45 11nm
B1* 41-45 36-40 15nm
B2* 36-40 36-40 15nm
B3* 31-35 36-40 11nm
B4* 26-30 36-40 11nm
C1* 41-45 31-35 15nm
C2* 36-40 31-35 15nm
C3* 31-35 31-35 15nm
C4* 26-30 31-35 11nm
D1* 41-45 26-30 15nm
D2* 36-40 26-30 15nm
D3* 31-35 26-30 15nm
D4* 26-30 26-30 11nm

Comparing proposal #3 and original playbook matrix: