Point of Decision Prompt Campaign

1
Point of Decision Prompt Campaign

• January 17, 2005
• And Ongoing

2
What Was Done

• Point of decision prompt (PODP) signs encouraged
  employees to choose the stairs instead of the
  elevator
• Employee stair, elevator, ramp, and building door
  use was monitored using infrared sensors and
  direct observation (via security camera images)
• Baseline data on lobby traffic were collected
  prior to posting the PODP signs and 12 weeks of
  campaign data were collected after posting the
  signs

3
Medtronic Data Collection Points
Visitor / Employee Entrance
Employee Entrance
Elev.
Elev.
?
?
(building)
(building)
?
?
?
(stair)
(stair)
?
(ramp)
(ramp)
Three infrared sensors in each lobby
?
sensor location
door
Parking Ramp
4
Infrared Sensor Data Problems

• For technical reasons there was no sensor at the
  elevator door. Elevator use was to be calculated
  by subtracting stair and ramp use from building
  door use
• Sensors registered multiple data captures per
  second between 10 am and midnight but not at
  other times of the day
• Door movement greatly affected sensor counts
• Due to sensor counts of door movement, we were
  unable to estimate elevator use

5
Sensors Registered Multiple Captures Per Second

• Between midnight and 10 am a sensor only recorded
  one data capture per second. Between 10 am and
  midnight a sensor recorded up to 5 data captures
  per second.
• This pattern occurred on all sensors, 7 days a
  week, from the first to the last day of data
  collection.
• We were unable to ascertain the reason for the
  multiple data captures.

6
Visitor/Employee Building SensorNumber of
Seconds with Recorded Activity
7
Door movement greatly affected sensor counts

• Between 7 and 9 am the building entrance door was
  propped open for the employee entrance but not
  for the visitor/employee entrance
• Building door counts are greatly reduced between
  7 to 9 am for the employee entrance

8
Visitor/Employee EntranceSensor Counts during
PODP Campaign by Time of Day
Building door NOT propped open
Data are from 1/17/2005 to 4/14/2005 excluding
1/28/2005 (technical problems) and 2/21/2005
(holiday)
9
Employee EntranceSensor Counts during PODP
Campaign by Time of Day
Building door propped open Notice building counts
are reduced
Data are from 1/17/2005 to 4/14/2005 excluding
1/28/2005 (technical problems) and 2/21/2005
(holiday)
10
Unable to Estimate Elevator Use

• Because of heavier traffic through the building
  door, the counts were reduced more than were the
  counts for stair or ramp doors which had less
  traffic
• Stair plus ramp use counts were about equal to or
  slightly greater than building door counts
• We could not estimate elevator use by subtracting
  stair and ramp counts from building counts

11
PODP Campaign Evaluation

• Because of problems with the sensor data, data
  from the security cameras were used to evaluate
  the effectiveness of the
• PODP campaign.

12
Camera Data

• Data were collected from the employee entrance
  only.
• Data were collected for elevator, stair, and ramp
  use. Traffic through the building door was
  calculated as the sum of elevator, stair, and
  door traffic.
• Data were collected weekdays between the hours of
  6 to 10 am and between 3 to 7 pm.

13
BaselineEmployee Entrance Camera Counts by Time
of Day
Almost twice as many people used the stairs as
used the elevator before the PODP campaign
Baseline consists of one week of data
(1/10/2005 1/14/2005) prior to PODP campaign.
14
PODP Campaign Employee Entrance Camera Counts
by Time of Day
Data are from 1/17/2005 to 4/8/2005 excluding
2/21/2005 (holiday)
15
Increase in Use of Stairs

• There was increased choice to use stairs between
  8 to 9 am and between 4 to 6 pm
• Full increase in stair use didnt occur until the
  second week of the PODP campaign
• Ramp use stayed constant. The PODP campaign did
  not reduce ramp use.

16
Stair Use as Percent of People Taking Either
Stairs or Elevator by Time of Day
Increase in choice to use stairs rather than
elevator
17
Stair Use as Percent of People Taking Either
Stairs or Elevator by Week
Baseline Rate
18
PODP Campaign Results

• Baseline 64.9 chose stairs and 35.1 chose
  elevator.
• PODP campaign, weeks 2 through 11, 69.7 chose
  stairs and 30.3 chose elevator.
• (35.1 - 30.3) / 35.1 13.7 of elevator users
  at baseline switched to stair use during the PODP
  campaign.