Archived Data Management System

1
Archived Data Management System

• Study Advisory Committee Meeting
• May 14, 2003

2
Preliminary Data Analysis

• 2002 AADT Estimation
• TRIMARC
• ARTIMIS

3
Procedure

• Data extraction and screening
• Applied quality check criteria used in mobility
  monitoring study by TTI
  
• Problem encountered The rounding of occupancy
  in the segment file created problems.
  Potential solution Eliminating the vehicle
  length criteria

4
Procedure (contd)

• Summarize 15min volume to hourly volume
• If there are at least 2 15min records, calculate
  hourly volume by adding or extrapolating
  
• Otherwise, mark hourly volume as missing/null.

5
Procedure (contd)

• Summarize hourly volume to daily volume
• If there are 24 hourly volume records for a day,
  add them.
  
• If there are 18-23 hourly volume records for a
  day, impute the hourly volumes for those
  missing hours based on hourly distribution at
  the same site throughout the year, and then add
  them.
• Otherwise, mark the day as missing/null.

6
Procedure (contd)

• For each site, screen out daily volumes on those
  days of week based on 1.5s from the day-of-week
  mean.
  
• Calculate day-of-week volume distribution using
  yearly average for each day of week.

7
Procedure (contd)

• Calculate AADT using the AASHTO formulation (I)
• where
• VOL daily traffic for day k, of day-of-week i,
  and month j
  
• i day of the week
• j month of the year
• k 1 when the day is the first occurrence of
  that day of the week in a month, 4 when it is the
  fourth day of the week.
  
• n the number of days of that day of the week
  during that month (usually between 1 and 5,
  depending on the number of missing data).

8
Quality Control Criteria
9
Quality Control Criteria (contd)
10
Quality Control Criteria (contd)
11
Data Screening Summary
12
Data Screening Summary (contd)
13
TRIMARC Data Quality
14
ARTIMIS Data Quality
15
Alternative Methods

• Calculate Monthly ADT (MADT) and multiply it with
  the monthly factor to get AADT (II)
  
• Estimate MADT using two-weeks of good data
  (with minimum 15min records marked as
  missing/null) and multiply it with the monthly
  factor to get AADT (III)

16
2003 Monthly Factor (Division of Planning, KYTC)
17
TRIMARC Summary (Method I)
18
TRIMARC Summary (Method II)
19
TRIMARC Summary (Method II) (contd)
20
TRIMARC Summary (Method II) (contd)
21
TRIMARC Summary (Method II) (contd)
22
TRIMARC Summary (Method II) (contd)
23
ARTIMIS Summary (Method I)
24
ARTIMIS Summary (Method II)
25
ARTIMIS Summary (Method II) (contd)
26
ARTIMIS Summary (Method II) (contd)
27
ARTIMIS Summary (Method II) (contd)
28
AADT Estimates at Sample Sites
29
Observations

• Data availability varied by sites in 2002.
• The rounding of occupancy data may have caused
  large amount of data being screened out by
  vehicle length criteria, which were hence dropped
  when processing ARTIMIS data.

30
Observations (contd)

• Generally, the AADT estimates obtained from
  method II are closer to the State figure when
  more valid days are present for a month.
  
• Significant differences exist between AADT
  estimates obtained using monthly factors for all
  week, weekday, and weekend (method III).