Sprinkler Reliability

1

Sprinkler Reliability and Performance Dr. John
R. Hall, Jr. National Fire Protection
Association May 2008
2
Overview of presentation

• Sprinkler statistics for decisions
• Recent changes in NFIRS data
• Reliability
• Effectiveness when operating
• Performance reliability Reliability x
  Effectiveness
• Side effects
• Cost affordability
• Cost-effectiveness
• System alternatives comparisons

3
Recent changes in NFIRS

• NFIRS
• Version 4.1 (pre-1999) and 5.0 (1999 on)
• Presence of automatic extinguishing equipment
• Type of automatic extinguishing equipment
• Operation vs. not vs. fire too small to activate
• Effectiveness

4
Sprinkler performancecoding options before 1999

• NFIRS version 4.1
• Automatic extinguishing equipment
• Operated
• Should have operated but did not
• Fire too small to require operation
• No equipment present
• Unclassified
• Unknown

5
Sprinkler performancecoding options since 1999

• NFIRS version 5.0
• Distinguish sprinklers from other automatic
  extinguishing equipment
• Distinguish wet, dry and other sprinklers
• Distinguish presence and operation from
  effectiveness
• Reasons for failure to operate or ineffectiveness
• Number of sprinklers operating

6
How we analyzed the new data

• Distinguish sprinkler statistics from statistics
  for all automatic extinguishing equipment.
• Check consistency between type of unsatisfactory
  performance and reasons for unsatisfactory
  performance. Recode as necessary.
• Analyze using percentages to avoid converted
  version 4.1 data

7
Recoding inconsistent data

• If Performance Not Effective
• And Reason System Shut Off
• Change to Performance Failed to Operate
• If Performance Not Effective
• And Reason Not in Area of Fire
• Change to Presence No

8
Recoding inconsistent data

• If Performance Failed to Operate
• And Reason Not Enough Agent OR Agent Did Not
  Reach Fire
• Change to Performance Not Effective
• If Performance Failed to Operate
• And Reason Not in Area of Fire
• Change to Presence No

9
Type of AES equipment

10
Type of AES equipment reported
11
Sprinklers vs. other AES equipment, by property
type
12
Operational reliability

13
Primary findings regarding operational reliability

• With pre-1999 data, we estimated 16 of
  sprinklers did not operate when they should,
  excluding fire too small to activate
• With newer data and no recoding, the estimate is
  18
• With newer data and recoding, estimate 7 for
  sprinklers only 93 operational reliability
• Majority (55) of reported fires are too small to
  activate equipment

14
Failure to operate when sprinklers present and
fire not too small
15
Reasons for sprinkler failure to operate, for all
properties
16
Can operational reliability be increased?

• Avoid system shut-offs
• Provide appropriate maintenance
• Avoid mismatch between property use and choice of
  systems
• Then based on reasons for failure to operate, the
  operational reliability could rise to 99½
• Which is comparable to the Australian data,
  collected where systems receive more consistent
  and rigorous oversight

17
Primary findings regarding effectiveness

• Effectiveness of operating sprinklers is 97
• Overall performance reliability operational
  reliability times effectiveness if it operates
  for sprinklers is 90

18
Percent of operating sprinklers that were deemed
effective
19
Reasons why operating sprinklers were deemed
ineffective
20
Percent of sprinklers that operated and were
deemed effective
21
Other effectiveness measuresImpact of sprinklers

• When sprinklers are present in the fire area in
  buildings that are not under construction
• Fire deaths per 100 fires are reduced by at least
  57, depending on property use
• Property damage per fire is reduced by one-third
  to two-thirds (34-68)
• Likelihood of flame damage beyond the room of
  origin goes down from 43 to 11

22
Summary of main findings

• Sprinklers fail to operate 7 of time
• If you can avoid system shut-offs, lack of
  maintenance and mismatched systems, this should
  drop to ½ -- comparable to the Australian data
  collected for systems with more consistent and
  rigorous oversight
• Effectiveness of operating sprinklers is 97
• Overall performance reliability operational
  reliability times effectiveness if it operates
  for sprinklers is 90 93 x 97

23
Summary of main findings

• Performance can be distinguished for dry pipe
  sprinklers
• Performance can be distinguished for automatic
  extinguishing equipment other than sprinklers
• Performance can be distinguished by property type
• All these new types of possible analyses offer
  potential guidance for research or performance
  improvement programs

24
Other measures for decisions

• Side effects water release with no fire. NFPA
  has new NFIRS data and plans for special study to
  amplify it.
• Cost Fire Protection Research Foundation has
  current project to obtain systematic data
• Cost affordability Argument primarily analogy
  with other building features. Past studies
  suggest that is not how people frame issue.
• Cost-effectiveness Rarely see quantification of
  costs and effectiveness of alternative system
  designs, except national policy analysis studies

25
Do we have complete FAQs?

• Every measure listed here is the subject of
  Frequently Asked Questions from decision-makers,
  opinion makers, and the public
• New NFIRS codes give us better answers for some
• New studies underway will give us better answers
  for more
• But we still dont know as much as we need to
  know about how various groups frame the issues
  and what information they need and want to make
  their decisions.