WP5 - Estimated Time of Arrival: verification of the F-Man approach and identification of effects on fleet management Athens -Greece 23-25 Sept 2004 National.

Slides:



Advertisements
Similar presentations
* Course Outline * EDP ( Electronics Data Processing ) * DOS (Disk Operating System ) * GW-Basic ( Computer Programming Language) * Windows ( Computer.
Advertisements

Naïve Bayes. Bayesian Reasoning Bayesian reasoning provides a probabilistic approach to inference. It is based on the assumption that the quantities of.
UGDIE PROJECT MEETING Athens September WP6 – Assessment and Evaluation  Evaluation Results  Deliverable D 13 (first release issued and delivered.
Session 8b Decision Models -- Prof. Juran.
6-1 Introduction To Empirical Models 6-1 Introduction To Empirical Models.
Confidence Intervals, Effect Size and Power
Chapter 9 Tests of Significance Target Goal: I can perform a significance test to support the alternative hypothesis. I can interpret P values in context.
WP5 - Estimated Time of Arrival: verification of the F-Man approach and identification of effects on fleet management Athens -Greece Sept 2004 National.
WP3- Development: Milestone M3 Lisbon 30 January 2004 National Technical University of Athens.
Class Handout #3 (Sections 1.8, 1.9)
Introduction to Hypothesis Testing
Figure 6-1: Transport Request Data Entities: CS: Customer OM: Order manager OPM: Operational manager FM: Fleet Manager MM: Maintenance manager OBT: On.
Tests of significance Confidence intervals are used when the goal of our analysis is to estimate an unknown parameter in the population. A second goal.
Software Quality Control Methods. Introduction Quality control methods have received a world wide surge of interest within the past couple of decades.
UGDIE PROJECT MEETING Bled September WP6 – Assessment and Evaluation Evaluation Planning  Draft Evaluation plan.
Statistics 101 Class 9. Overview Last class Last class Our FAVORATE 3 distributions Our FAVORATE 3 distributions The one sample Z-test The one sample.
8-1 Copyright ©2011 Pearson Education, Inc. publishing as Prentice Hall Chapter 8 Confidence Interval Estimation Statistics for Managers using Microsoft.
Copyright ©2011 Pearson Education 8-1 Chapter 8 Confidence Interval Estimation Statistics for Managers using Microsoft Excel 6 th Global Edition.
QMS 6351 Statistics and Research Methods Chapter 7 Sampling and Sampling Distributions Prof. Vera Adamchik.
MR2300: MARKETING RESEARCH PAUL TILLEY Unit 10: Basic Data Analysis.
Data Analysis Statistics. Inferential statistics.
MOS Performance MOS significantly improves on the skill of model output. National Weather Service verification statistics have shown a narrowing gap between.
BCOR 1020 Business Statistics Lecture 20 – April 3, 2008.
WP4- Integration Kranjka Gora - Slovenia April 2004 National Technical University of Athens.
Model Calibration and Model Validation
Chapter 9: Introduction to the t statistic
Airline On Time Performance Systems Design Project by Matthias Chan.
DUMMY CLASSIFICATION WITH MORE THAN TWO CATEGORIES This sequence explains how to extend the dummy variable technique to handle a qualitative explanatory.
1 Doing Statistics for Business Doing Statistics for Business Data, Inference, and Decision Making Marilyn K. Pelosi Theresa M. Sandifer Chapter 15 The.
Statistical hypothesis testing – Inferential statistics I.
Training. Why Train? skills and knowledge needed by new staff update skills of old staff assure conformity to standards teach the proper use of SQA procedures.
Software Reliability Growth. Three Questions Frequently Asked Just Prior to Release 1.Is this version of software ready for release (however “ready” is.
Review of Basic Statistics. Definitions Population - The set of all items of interest in a statistical problem e.g. - Houses in Sacramento Parameter -
Great Basin Verification Task 2008 Increased Variability Review of 2008 April through July Period Forecast for 4 Selected Basins Determine what verification.
N By: Md Rezaul Huda Reza n
1 DATA DESCRIPTION. 2 Units l Unit: entity we are studying, subject if human being l Each unit/subject has certain parameters, e.g., a student (subject)
Statistical Analysis A Quick Overview. The Scientific Method Establishing a hypothesis (idea) Collecting evidence (often in the form of numerical data)
1 CSI5388: Functional Elements of Statistics for Machine Learning Part I.
Chapter 8 Introduction to Hypothesis Testing
Individual values of X Frequency How many individuals   Distribution of a population.
Fundamentals of Data Analysis Lecture 9 Management of data sets and improving the precision of measurement.
Data Analysis (continued). Analyzing the Results of Research Investigations Two basic ways of describing the results Two basic ways of describing the.
Developed at Utah State University Dept of Engr & Tech Educ — Materials and Processes 5.6 calculate the mean and standard deviation of.
AP STATISTICS LESSON 10 – 2 DAY 1 TEST OF SIGNIFICANCE.
Caminhos de Ferro Portugueses E.P. European Datacomm NV IVU Traffic Technologies AG National Technical University of Athens Prometni Institut Ljubljana.
PCB 3043L - General Ecology Data Analysis. OUTLINE Organizing an ecological study Basic sampling terminology Statistical analysis of data –Why use statistics?
Lecture 16 Section 8.1 Objectives: Testing Statistical Hypotheses − Stating hypotheses statements − Type I and II errors − Conducting a hypothesis test.
Marion Hughes Sociology 391 Spring Q. 110: How many days out of the past 30 have you used marijuana?  0  1-5  6-10    21+ Recoded.
INTRODUCTORY STUDY : WATER INDICATORS AND STATISTICAL ANALYSIS OF THE HYDROLOGICAL DATA EAST OF GUADIANA RIVER by Nikolas Kotsovinos,P. Angelidis, V. Hrissanthou,
Chapter 10 Verification and Validation of Simulation Models
1 Introduction to Transportation Systems. 2 PART II: FREIGHT TRANSPORTATION.
Logic and Vocabulary of Hypothesis Tests Chapter 13.
Chap 8-1 Chapter 8 Confidence Interval Estimation Statistics for Managers Using Microsoft Excel 7 th Edition, Global Edition Copyright ©2014 Pearson Education.
Inferential Statistics Inferential statistics allow us to infer the characteristic(s) of a population from sample data Slightly different terms and symbols.
1 Module One: Measurements and Uncertainties No measurement can perfectly determine the value of the quantity being measured. The uncertainty of a measurement.
History & Research Research Methods Unit 1 / Learning Goal 2.
 Simulation enables the study of complex system.  Simulation is a good approach when analytic study of a system is not possible or very complex.  Informational,
Quality Control Copyright © 2015 McGraw-Hill Education. All rights reserved. No reproduction or distribution without the prior written consent of McGraw-Hill.
WEPS Climate Data (Cligen and Windgen). Climate Data ▸ Climate Generation by Stochastic Process A stochastic process is one involving a randomly determined.
Copyright © 2013 Pearson Education, Inc. Publishing as Prentice Hall Statistics for Business and Economics 8 th Edition Chapter 9 Hypothesis Testing: Single.
Building Valid, Credible & Appropriately Detailed Simulation Models
15 Inferential Statistics.
Unit 5: Hypothesis Testing
PCB 3043L - General Ecology Data Analysis.
Measures of Central Tendency
Dr. Cleveland, This file contents 4 tracks poster slides
Chapter Nine Part 1 (Sections 9.1 & 9.2) Hypothesis Testing
Jeroen Pannekoek, Sander Scholtus and Mark van der Loo
Chapter 9 Hypothesis Testing: Single Population
Skills 5. Skills 5 Standard deviation What is it used for? This statistical test is used for measuring the degree of dispersion. It is another way.
Presentation transcript:

WP5 - Estimated Time of Arrival: verification of the F-Man approach and identification of effects on fleet management Athens -Greece Sept 2004 National Technical University of Athens

CONTENT National Technical University of Athens 1.Verification of the ETA approach > Method #1 – Using statistical records > Method #2 - Through F-man pilot 2.Identification of the ETA effects in fleet management

Ideal Data required National Technical University of Athens Data Requirements for verification of ETA approach Statistical data (e.g. for 5 years) from international transport cases where  scheduled arrival/departure times as well as  actual arrival/departure times should be recoded in all itinerary stations (the knowledge of the reason of delay could be also quite useful).

Data availability National Technical University of Athens Available data: 1. National corridor: Alcantara–Lisboa–Entrocamento–Pamhilhosa-Gaia–Leixoes. Available records: 287 Period: Jan 2002 to March Data adequate for elaboration: Three international corridors (portuguese part) (I) Bobadela (PT)- Irún (SP) (II) Alcantara (PT)-Vigo (SP) (III) Vigo (SP)- Alcantara (PT) Available records: 340 Period: Jan 2002 to March Data adequate for elaboration: 266

Data elaboration by use of an on-purpose software tool National Technical University of Athens For the analysis of appropriate information from the raw data provided by the Portuguese railways in electronic form a dedicated software program was developed. This software searches automatically the numerous records provided to identify scheduled and actual arrival and departure times for user-selected itineraries (described through origin and destination stations), then calculates the relevant delay and stores the data in Excel format to allow further processing.

Verification Process National Technical University of Athens Based on the above data sets (although a great discrepancy exists between the ideally required and the available information) two verification approaches were identified: 1.A verification check based solely on statistical methods where the Alcantara–Leixoes data set were used. The data sample was split in two parts. The first part was used for the calibration of the ETA function while the second one was used for the verification test 2.A verification check based on data/information collected through the F-MAN pilot runs where the provided statistical information from Bobadela-Irún, Alcantara-Vigo and Vigo-Alcantara were used to calibration of ETA functions

Verification Process National Technical University of Athens Test-a Test-b

ETA method verification using statistical data for the corridor Alcantara-Leioxoes National Technical University of Athens  The calibration of ETA functions was performed by use of 2003 records.  The testing of ETA predictability was performed by use of 2004 records  The station of observations used was Entrocamento (at 113 km distance from Alcantara)  The destination station used was Leixoes (at a 364 km distance from Alcantara).

Calibration of ETA functions (based on 2003 data for Alcantara-Leioxoes Corridor) National Technical University of Athens Area 1 Area 2 Area 3 Area 4

Early departure from Entrocamento station (prior to the scheduled departure time) (Delay at Origin<0) National Technical University of Athens

On time departure from Entrocamento station (Delay at Origin=0) National Technical University of Athens

Delayed departure from Entrocamento (Delay >0) National Technical University of Athens 82%

Frequency & Cumulative distribution of actual delay at Leixoes fir delay at Entrocamento>0 National Technical University of Athens

Frequency & Cumulative distribution of actual delay at Leixoes fir delay at Entrocamento>0 National Technical University of Athens

Success and Failure Types SUCCESS II: Wagon REJECTED for USE and NOT ARRIVED ON TIME SUCCESS I: Wagon accepted for USE and ARRIVED ON TIME FAILURE I: Wagon accepted for USE and NOT ARRIVED ON TIME FAILURE II: Wagon REJECTED for USE and ARRIVED ON TIME

Testing ETA predictability using statistical information (Time margin=0,5 hours)- Sample 2004 Testing predictability method for time margin of delay at Leixoes=0,5 hours

Testing ETA predictability using statistical information (Time margin=1 hour)-Sample 2004 Testing predictability method for time margin of delay at Leixoes=1 hour

Testing ETA predictability using statistical information (Time margin=2 hours) –Sample 2004 Testing predictability method for time margin of delay at Leixoes= 2 hours

Testing ETA predictability using statistical information (Time margin=0,5 hours)- Sample 2002 Testing predictability method for time margin of delay at Leixoes=0,5 hours

Results of testing ETA predictability (I) The results of the above analysis indicate high success rates (“Use” cases subset) of ETA forecast, for all time margins. However, the subset of “Reject” cases is systematically under estimated. The reason for this is that train delays in 2004, are significantly lower than those of 2003 (see Figure 14). This probably happens because the operating conditions on the specific corridor favorably evolve.

Results of testing ETA predictability with the sample of 2002)

Results of testing ETA predictability with the sample of 2004

Reject cases underestimation Average delay of arrival at Leixoes (destination station) per month

ETA method verification using data from pilot runs  The calibration of ETA functions was performed by use of 2003 records for the portuguese part of the international corridor Bobadela- Irun.  The station of observations used was Entrocamento (at 106km distance from Bobadela)  The destination station used was Vilar Formoso (at a 327 km distance from Bobadela).  The testing of ETA predictability will be performed based on information from SMS messages produced during the pilot run process

Calibration of ETA functions (pilot)

Early departure from Entrocamento station (prior to the scheduled departure time) (Delay at Origin<0)

On time departure from Entrocamento station (Delay at Origin=0)

Delayed departure from Entrocamento (Delay >0)

Frequency & Cumulative distribution of actual delay at Vilar Formoso fir delay at Entrocamento>0

Testing predictability by use of data from the pilot run the ongoing process of ETA verification through Concerns this step. The procedure is similar to the one presented in the case of using historical data from national corridor Alcantara-Leixoes but instead of the 2004 statistical sample information from SMS messages will be used.

Understanding ETA effects: Simplified examples (Example #1)

Understanding ETA effects: Simplified examples (Example #2)