RECURRENT-EVENTS DATA ANALYSIS FOR PRODUCT REPAIRS, DISEASE RECURRENCES, AND OTHER APPLICATIONS

The Albany, NY Chapter of the American Statistical Association and

the University at Albany’s School of Public Health

announce a one-day course on Saturday, April 5, 2014 from 9-4

at the Gilda George Education Center

on the UAlbany’s East Campus, (in East Greenbush)

 

RECURRENT-EVENTS DATA ANALYSIS FOR PRODUCT REPAIRS, DISEASE RECURRENCES, AND OTHER APPLICATIONS

 Instructor Wayne Nelson, consultant, WNconsult@aol.com

OVERVIEW.

Most reliability and survival data analyses deal with data with one event for each sample unit, the end of life.  However, in many applications, sample units have repeated events, such as repairs of products, recurrences of tumors, consumer purchases on the Internet, remarriages, and reincarcerations.  This course presents models and analyses for such recurrent-events data, which do not yet appear in most texts.  This course is intended for practitioners who analyze recurrent events data.  Needed background is a one-year basic statistics course. The course is based on Wayne Nelson's book RECURRENT EVENTS DATA ANALYSIS FOR PRODUCT REPAIRS, DISEASE RECURRENCES, AND OTHER APPLICATIONS, published by ASA-SIAM in 2003, www.siam.org/books/sa10/.  

 

CONTENT.

1. INTRODUCTION TO RECURRENT EVENTS DATA AND APPLICATIONS

   This chapter describes recurrent events data on a sample of units from a population and the information sought from such data.  Such data are illustrated with three data sets: transmission repairs on cars, bladder tumor recurrences, and births of children to statisticians.  The first two sets contain exactly observed event and censoring times; the third contains interval (grouped) data.  Informative plots for displaying such data are presented.  This chapter also lists other applications and overviews the book.

   2.   POPULATION MODEL, MCF, AND BASIC CONCEPTS   This chapter presents the nonparametric population model for such recurrent-events data.  A simple stochastic process model, it consists of a cumulative history function for each population unit.  These population functions are summarized with the population Mean Cumulative Function (MCF) for the “cost” or number of recurrences.  The MCF yields most information of interest in applications, for example, the number of transmission repairs on warranty and the tumor recurrence rate over time.  The model extends to continuous history functions and to left censored and interval data, and is illustrated with other applications.

   3.   ESTIMATE OF THE MCF FOR EXACT DATA   This chapter presents the basic

   nonparametric estimate of the MCF, its plot, and the plot’s interpretation.  It shows how to calculate and plot the MCF estimate for exact data (exact values of event and censoring times).  The MCF estimate is illustrated with the transmission and bladder tumor data.  This chapter shows how to interpret a plot to get, among other information, 

· the behavior of the recurrence rate (increasing or decreasing) as the population ages, important for product burn-in, overhaul, and retirement decisions,

· a prediction of the number or cost of recurrences for sample units in a future time period,

· an estimate of the average number or cost of recurrences up to a specified age, such as warranty age or at infinity,

· a comparison of data sets from different product designs, medical treatments, populations, etc.,

· unexpected useful information, a great advantage of data plots.

This includes a survey of computer programs that calculate and plot the MCF estimate.   A technical section explains the underlying assumptions and properties of the MCF estimate and typical difficulties in applications.

Figure 3.2  Mean cumulative number of bladder tumor

                         recurrences of patients under Placebo treatment .

 

4.   CONFIDENCE LIMITS FOR THE MCF   This chapter presents approximate

confidence limits for the MCF for exact data.  They are illustrated with the transmission and tumor data.  This chapter surveys computer programs that calculate and plot the MCF estimate and confidence limits.  Included are the underlying assumptions and properties of the limits.

5.   ANALYSIS OF DATA WITH A MIX OF EVENTS   This chapter deals with data with a

mix of events, for example, a product may fail from a number of causes.  Usually one seeks to estimate the MCF

· for all events combined,

· separately for a particular event (say, a particular failure mode),

· for certain combinations of events (say, all types of failures on a particular subsystem),

· for the population when certain events are eliminated (say, through a redesign that eliminates failure modes).

These estimates are illustrated with data on subway car motors and naval turbines.

6.   ESTIMATE OF THE MCF FOR INTERVAL DATA   This chapter provides an

estimate of the MCF for interval data, where event and censoring times are grouped into intervals.  The estimate is illustrated with the childbirth data, which includes a comparison of the MCFs of men and women.

 

7.   COMPARISON OF SAMPLES   This chapter provides confidence limits and a plot

to compare two sample MCFs.  This is illustrated with the transmission and tumor data sets. The chapter surveys computer programs that calculate and make the plot.  A technical section explains the underlying assumptions and properties of the method.

Figure 7.2C   Difference (Placebo-Thiotepa) of MCFs of

bladder tumor recurrences and 95% limits.

 

8.   SURVEY OF FURTHER METHODS   This chapter surveys parametric methods

for analysis of recurrence data.   Topics include

· the Poisson and nonhomogeneous Poisson models, data analyses, and software,

· reliability growth models, data analyses, and software,

· renewal models and data analyses,

· models and data analyses for a single observed unit, rather than a sample of units,

· Poisson models with covariates, data analyses, and software,

· Cox model for recurrent events with covariates, data analyses, and software.

 

REFERENCES

All the data sets in the book are available in an Excel file at www.siam.org/books/sa10/BookData.xls          

 

 

 

 

 

SCHEDULE.

8:30-9:00  Registration

9:00-10:30   Introduction to Recurrent Events Applications and Population Model and Concepts.

10:30-10:45  Break

10:45-12:00  MCF Estimate, Plot, Confidence limits, and Interpretation

12:00-1:00  Lunch on your own with Wayne Nelson

1:00-2:45  Analysis of a Mix of Events and Interval Age Data

2:45-3:00  Break

3:00-4:30  Comparison of Samples and Survey of Additional Topics

4:30  End.

 

For information on course content, contact Wayne Nelson at WNconsult@aol.com .                        

 

COST. 

The course fee is $90 for non-students and $80 for full-time students, which includes the text by Wayne Nelson and a supplemental handout.  If you own the text, the fee is $10 for non-students and $5 for students, which includes a supplemental handout.  Payment by credit card or check accepted. An additional fee will be applied for credit card payment. See the registration section for more information.

 

REGISTRATION.

• The registration website is:  https://docs.google.com/forms/d/10cdIkueqOyI4t2s-x3RyLycQP6L-MVkTa62g07gMHPs/viewform

• Registration will not be considered complete until the full payment has been received.  

• Credit card payments can be made online at the time of registration.  The following additional fees will apply to all credit card payments (to cover processing fees): a$5 for registrations including the textbook, and a $2 fee for registrations not including the text book.

• Checks should be made payable to the ASA Albany, NY Chapter and mailed to: The Albany, NY Chapter of ASA, PO Box 5784, Colonie Center Station, Albany, NY 12205-5784.

• Registration and payment confirmation will be emailed to all registrants.

   

  For questions regarding registration, contact Laura Morris at amstat.albanyny@gmail.com.  

   

   

  INSTRUCTOR. 

  Dr. Wayne Nelson is a leading expert on reliability and accelerated test data analysis.  Formerly with General Electric Research & Development for 24 years, he now consults on and teaches engineering applications of Statistics for many companies, professional societies, and universities.  For his contributions to reliability data analysis and accelerated testing, he was elected a Fellow of the Amer. Statistical Assoc. (ASA), the Inst. of Electrical and Electronics Engineers (IEEE), and the Amer. Soc. for Quality (ASQ).  He also  authored two well-known Wiley books ACCELERATED TESTING and APPLIED LIFE DATA ANALYSIS.  Among his 120+ publications, he received the Brumbaugh, Wilcoxon, and Youden Prizes of ASQ and 9 outstanding presentation awards from ASA.  ASQ awarded him the 2003 Shewhart Medal for technical leadership and the 2010 Dorian Shainin Medal for contributions to accelerated testing.  In 2005, the IEEE Reliability Soc. conferred on him its most prestigious Lifetime Achievement Award for his many outstanding contributions to reliability and accelerated test data analysis and reliability education. 

   

   

  LOCATION, DIRECTIONS, and MAPS.

  The University at Albany's East Campus is located at 1 University Place Rensselaer, NY 12144, in East Greenbush, NY.  See map at http://mapq.st/1bjyIh8.  The room information will be distributed following registration.

   

  From most points of origin, easiest access is from Interstate 787 –

• If you are traveling to the area on the NYS Thruway, we recommend that you take Exit 23 and head North on Route 787.

• From Route 787, take the Rensselaer Exit for Rtes. 9 and 20 off of Rte. 787. This travels across the Hudson River on the Dunn Memorial Bridge

• At the end of the bridge, stay to the left, following signs to East Greenbush via Rtes. 9 and 20.

• Follow Columbia Turnpike (Routes 9 and 20) up the hill. At the fourth traffic light, make a left onto Discovery Drive.

• The University at Albany East Campus

   

  If you are coming from the east, along the Massachusetts Turnpike, or from the southeast, up the Taconic Parkway –

• Take exit onto I-90 Westbound, headed toward Albany and Buffalo.

• Stay on I-90 westbound to Exit 9 (NYS Rte 4). Exit and head South on Rte 4.

• About 1/4 mile S. on Rte. 4 is the first traffic light. Turn Right onto Rte 151.

• About a mile down Rte. 151 is the first traffic light.  Turn Left onto Sherwood. (there is no right turn at this light)

• Sherwood runs about 1/2 mile through a residential area and ends on "Columbia Turnpike" (Rtes 9 and 20). Turn Right onto Columbia Turnpike.  Continue past K-Mart on the left, and Burger King on the right.

• In about 1/2 to 3/4 mile immediately past a Cottman Transmission shop, there is a traffic signal; the entrance to the East Campus is on the right.  Turn right at the traffic light.

• Drive past the building under construction to the left and past Regeneron Pharmaceuticals on the right.

• A left turn leads into the SPH parking area.  SPH is in the large, buff-colored building (the "George Education Center") to your left as you pull into the parking lane.  Entry is through the glass atrium.

   

  GPS Directions

• Enter the following address into your GPS for the School of Public Health:

  1 University Place

  Rensselaer, NY 12144

• Please note- when entering the address in your GPS, the "city" field should be Rensselaer NOT Albany. If you search for the University at Albany in your GPS, it will bring you to the Main Campus in Albany. The course is not on that campus.  The course is on the East Campus in Rensselaer, NY. 

   

  Interactive Google Map of East Campus

• To obtain directions to the East Campus from your current location, please use UAlbany's Interactive Google Map. When choosing your destination, please select: Campus: East Campus
  Area: Academic
  Specific Area: Gilda George Education Center

   

  Printable Maps

• Visit the UALBANY SPH website at: http://www.albany.edu/sph/map_directions.php