曝光臺 注意防騙 網(wǎng)曝天貓店富美金盛家居專營店坑蒙拐騙欺詐消費(fèi)者

HANDLING PROCESS TIMES DATA 67
3
List of Tables
1.1 Aircraft Properties . . . . . . . . . . . . . . . . . . . . . . . . . 9
1.2 Form of the Available Data . . . . . . . . . . . . . . . . . . . . . 10
4
List of Figures
1.1 A Part of AIS Handbook . . . . . . . . . . . . . . . . . . . . . . 13
2.1 Gamma Q-Q Plot for shifted GHP Times . . . . . . . . . . . . . . 18
2.2 Histogram of Shifted GHP Data with bin-width=3.15 . . . . . . . 19
2.3 Histogram of Shifted GHP Data with Rearranged Bins . . . . . . 20
2.4 Quantile Regression  function . . . . . . . . . . . . . . . . . . . 21
2.5 Quantiles of GHP Durations of Year 2008 OP2 . . . . . . . . . . 23
2.6 Gamma Distribution Fit by Optimizing the Distance between Theoretical
and Sample Quantiles . . . . . . . . . . . . . . . . . . . 25
2.7 Quantile-Quantile Plot of the Shifted Sample Quantiles and Estimated
Gamma Quantiles . . . . . . . . . . . . . . . . . . . . . . 26
2.8 Quantile-Quantile Plot of the Theoretical Weibull Quantiles with
MLE parameters and the Sample Quantiles . . . . . . . . . . . . 28
2.9 Weibull Distribution Fit by Optimizing the Distance between Theoretical
and Sample Quantiles . . . . . . . . . . . . . . . . . . . 29
3.1 Graphical Representation of The 0.1 Quantiles of GHP Time of
Boeing 737-400(ICA) over OP Periods . . . . . . . . . . . . . . . 36
3.2 Graphical Representation of The 0.2 Quantiles of GHP Time of
Boeing 737-400(ICA) over OP Periods . . . . . . . . . . . . . . . 37
3.3 Autoregression Coefficients Graph for 0.1 Quantiles of GHP Times
over Lags . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38
5
3.4 Autoregression Coefficients Graph for 0.5 Quantiles (Median) of
GHP Times over Lags . . . . . . . . . . . . . . . . . . . . . . . . 39
6
Chapter 1
INTRODUCTION
In order to manage the processes on the day of operations, Royal Dutch Airlines
(KLM) has split up the aircraft processes into so called Building Blocks.
Each Building Block may consist of various sub-processes, but has one Process
Owner with whom the Operations Control Department has Service Level Agreement
(SLA) about quality and throughput times.
As a part of the SLA, each Process Owner delivers a forecast of the statistical distribution
for their Building Block processes for the coming schedule season. The
distributions are described by means of a sequance of percentiles. These statistics
are used by the schedule developers and by Operations Control to determine
whether a new schedule can meet the punctuality standards which are set by the
KLM Management. Moreover, during an operational schedule period, a check is
continuously performed to determine whether the actual process times still match
the forecasted statistics. Corrective measures may be taken if this is not the case.
The Ground Services Department is one of the Process Owners, and delivers statistical
distributions for the ground handling process at Amsterdam Schiphol Airport,
the Netherlands, and at all outstations. The statistics consist of a set of percentiles
based on historical data.
7
Until recently, a Gamma distribution was fitted on the given percentiles before
performing a simulation with the OpiuM (The performance model OPiuM is a
computer simulation model that is used by KLM to evaluate the performance and
robustness of a proposed flight schedule) model. Because this introduces an undesired
interpretation of the process statistics, it has been decided to replace the
Gamma distribution by a distribution consisting of line segments connecting the
percentiles. However, the distribution consisting of line segments appears not to
give realistic results when a schedule forecast is performed. It is suspected that this
is due to inaccuracies in the given statistics. A possible explanation is that some
categories have an insufficient number of observations from which to derive the
statistics.
This work will consist in developing a statistical model to provide a more accurate
forecast of the process statistics, based on historical data.
The observations have been made on different type of aircrafts on different routes
on operation plan periods since 2006. Each aircraft differs in size; therefore, the
standard ground handling process time changes for each group. Furthermore, the
process times depend on the type of the flight, which can be departure, turn-around,
or so on.
Table 1.1 shows the aircraft types we studied on and their capacities.
The form of the data we used is shown in Table 1.2.
In table 1.2, BB24 pln states the standard ground handling process times which
is determined by KLM Management. ”BB24 Act” is for actual observed process
　
中國航空網(wǎng) www.k6050.com
航空翻譯 www.aviation.cn
本文鏈接地址：航空資料31(10)