Building and Environment 51 (2012) 311e319

Contents lists available at SciVerse ScienceDirect

Building and Environment

journal homepage: www.elsevier.com/locate/buildenv

The adoption of fire safety management for upgrading the fire safety level of
existing hotel buildings
Ying-Yueh Chen a, b, *, Ying-Ji Chuang a, Chin-Hsing Huang a, d, Ching-Yuan Lin a, Shen-Wen Chien c
a
Department of Architecture, National Taiwan University of Science and Technology, Taiwan
b
Department of Fire Research and Advance, Taiwan Architecture and Building Center, Taiwan
c
Graduate School of Fire Science and Technology, Central Police University, Taiwan
d
Nanya Institude of Technology, Taiwan

a r t i c l e i n f o a b s t r a c t

Article history: The fire safety of existing buildings decreases over time. Based on factors such as space limitation and
Received 28 August 2011 high construction costs, upgrading hardware construction is difficult, especially in existing hotel build-
Received in revised form ings due to their 24-h operation. This paper proposes a simple fire safety evaluation system for existing
29 November 2011
multi-purpose hotel buildings. Fire safety management is the most important section in the evaluation
Accepted 3 December 2011
system, which includes fire prevention, and evacuation and mitigation strategies, accounting for a 54.95%
weighting. In this study, an empirical evaluation was conducted on 16 multi-purpose hotel buildings. Of
Keywords:
these test cases, five were determined to be unsafe. After reviewing the scores for each fire safety
Existing hotel buildings
Fire safety management measures
management plan factor, necessary improvement plans were proposed. This system will assist owners in
Fire safety evaluation system undertaking improved fire safety management measures with less hardware renewal.
Improvement plans Ó 2011 Elsevier Ltd. All rights reserved.

1. Introduction engineers submit suggestions for hardware improvements

according to the evaluation system. These evaluation systems are
Fire safety for existing buildings decreases over time. In order to put in place in conjunction with fire insurance and financial loan
ensure the safety of buildings, hardware upgrades and fire safety mechanisms, making it is easier to request property owners to
management measures are essential. However, due to space limi- undertake the recommended improvements.
tations and high construction costs [1], upgrading hardware In Hong Kong, upgrading building hardware is difficult in
construction is difficult for existing multi-purpose hotel buildings. existing high-rise buildings. The EB-FSRS is a simple evaluation
Therefore, it is necessary to consider the fire risks characteristic of system that quantifies the fire safety level of old high-rise buildings
this type of building. with a scoring system that prioritizes fire safety improvement
The fire risk characteristics of existing multi-purpose use works [3,11e14]. Fire safety management considerations are of
buildings are not only complex, but are particular to each building particular importance for old high-rise buildings [4,15], old high-
[2e4] due to the different use behavior and culture of the individual rise non-residential buildings [1] and existing karaoke establish-
owners. In Taiwan, owners tend to decorate without considering ments [16,17].
the fire safety of their buildings. Due to the high frequency of fire incidents at hotels in various
Since 1994, international experts have studied fire risk evalua- countries, hotels are listed as “high risk” buildings [18e23].
tion systems for buildings. In 1973, Gretener proposed the Points Because of their multi-purpose use, frequent access of non-specific
Scheme, which is used to compare the tolerance limits of danger personnel, and the low environmental sensitivity of entrants,
and the real danger facing buildings [5,6], and in 1994, NFPA101A hotels need to implement a fire safety management system to
and Watts Jr. presented the Fire Safety Evaluation System (FSES) of effectively guide individuals in proper fire evacuation procedures
the Life Safety Code [7e10]. This system is based on prioritizing in order to ensure their personal safety. In high-rise British hotels,
human safety with four evaluation dimensions. Fire prevention out of 78% of fire incidents, nearly one third involved individual
evacuation [24]; as a result, emergency response training and daily
fire fighting equipment maintenance helped to ensure human
* Corresponding author. Department of Architecture, National Taiwan University
safety during a disaster [25e27]. Fire prevention management
of Science and Technology, No.43, Sec.4, Keelung Rd., Taipei 106, Taiwan. Tel.: þ886
2 86676 111105; fax: þ886 2 86676397. training of housekeeping staff was also found to increase safety
E-mail address: [email protected] (Y.-Y. Chen). levels [28].

0360-1323/$ e see front matter Ó 2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.buildenv.2011.12.001
312 Y.-Y. Chen et al. / Building and Environment 51 (2012) 311e319

Taiwan’s hotels have increased their business by 25% on an 3. Fire safety management measures used in evaluation
annual basis [29] and now accommodate nearly 56 million tourists. systems for existing multi-purpose hotel buildings
Consequently, Taiwan’s government has listed existing hotels in
need of priority improvements. Due to their 24-h operation, it is not This paper adopted human behavior as the basis for considering
easy to upgrade the hardware of these buildings. three safety issues [18]: users’ behavior characteristics, fire char-
This study proposes a fire management plan that serves as an acteristics and building characteristics, in order to address the
important risk indicator when evaluating the overall fire safety of causes of fires in relevant case studies, as well as to determine the
existing multi-purpose hotel buildings [30], and by which the fire safety risks of user behavior. Statistical analysis, research
safety levels for the reference of fire insurance and financial loan discussion and expert interviews are conducted to acquire risk
applications can be classified. In this paper, reference is made to the evaluation factors to construct a risk evaluation system for this
Fire Safety Guidelines of European Hotels, published by the EU study (shown in Fig. 1).
HOTREC in 2010 [31], in order to propose a simple evaluation for
existing multi-purpose use hotels. It is expected that the imple- 1. The three strategies (Fire Prevention strategy, Evacuation and
mentation of this evaluation technique would improve the fire Mitigation strategy, and Fire Control and Resistance strategy)
safety of buildings which would otherwise be difficult to update. were selected as key risk factors and evaluation of importance
and fitness of these factors were evaluated by 50 experts (from
the fields of construction, fire protection, hotel management,
2. Review of Taiwan’s building and fire codes for existing and insurance); the final key factor and its weight as well as
buildings require qualified factors are shown in Section 3.1.
2. In consideration of use behaviors, five risk characteristics are
Buildings in Taiwan are reviewed and evaluated by strict rules related. The characteristics of type of business, short-term
and regulations before being granted licenses, while hotel buildings occupants, fire management plans, enhancement of service
must be registered with the tourism authorities of the government facilities, and fire prevention and training issues are used to
before opening for business. The fire safety of existing hotel help to understand the utilization of reduced value at risk,
buildings is inspected and reported to the Building Administration URred, to calculate the score values of fire strategies as shown in
and Fire Department [32,33] to ensure the fire safety functions of Section 3.2.
the buildings. 3. Safety level is decided by a five-level and based on bench-
However, conflicts can exist in the fire safety maintenance of marking, each strategy is scored (R1, R2, R3) and ranked for the
buildings between two departments due to different duties; for level of total scores (Rg). The lowest level of each strategy is
example, the fire door (fire rolling shutter), when being closed or ranked as Level R. If the score of the require qualified factors
shut down, must be linked with a thermal-type fire detector or fails to reach 1/2 scores, it will then be downgraded by one
sprinkler. But, in an inspection of the performance of the fire door level as described in Section 3.3.
(fire rolling shutter), only the fire detector or sprinkler will be 4. According to the evaluation result of Level R and the score of
checked, and not whether the fire door (fire rolling shutter) itself each strategy (R1, R2, R3), a fire engineer can propose an
can work precisely in case of a fire. This is due to the fire door being improvement plan as well as to help to enhance the fire safety
the responsibility of the Building Department and the Fire of existing hotels.
Department being responsible for the detector; both of them are
independent and do not communicate their findings with each 3.1. Evaluation factors and qualified factors
other.
This a common type of problem for existing buildings. Although There is a direct correlation between a building’s characteristics
existing buildings may comply with legal requirements, there is and how a fire spreads, as well as the number of casualties and
a lack of overall fire safety evaluation as property owners do not amount of financial loss during a fire [28]. In terms of the evaluation
have sufficient professional knowledge to ensure the safety of the factors of hotel building fire safety, this study identified the eval-
building’s occupants. uation factors in nine investigation reports between 1980 and 2006
In order to ensure that building safety will not be influenced by evaluated by the NFPA [35e43] on fire incidents in hotel buildings
changes in rules and regulations, the Taiwanese government in Taiwan (Table 1).
enacted “The Rules of Hardware Updating of Evacuation Facilities The fire safety of buildings is determined by fire and smoke
and Fire Safety Equipment of Existing Buildings” in 2007 [34]; spread, evacuation safety, location of the fire and causes of fire.
however, due to the requirements of construction costs, space and Evaluation factors are identified according to the disaster-cause
operation for the update of building hardware, as well as the lack of ratio, as shown below:
the overall acceptance of fire prevention safety measures, this
legislation has been difficult to implement. In terms of the failure to Di;j þ Ii;j
Disaster-Cause Ratio : Ci;j ¼ Pj   (1)
comply with the relevant fire prevention rules and regulations, the
1 Di;j þ Ii;j
main reason is that government and property owners try to avoid
addressing the issue; however, a “wait and see” attitude is of no where, i refers to the four dimensions, j is the cause, D is the number
help when a problem occurs. of deaths, and I is the number of injured; while i ¼ 1e3, 1 is the fire
In order to resolve the issue, this paper has taken the perspec- and smoke spread, 2 is evacuation safety, 3 is the location of the fire,
tive of hotel owners (operators) and considered the difficulty of and 4 is the cause.
construction and spatial improvements for buildings. Without
interrupting operations, a risk evaluation procedure, which would 1. Disaster-cause dimension of fire and smoke spread in
be an important indicator of a fire management scheme for hotel buildings:
operators wanting to improve the safety level of their building, was (1) No fire barriers along the vertical pipes, stairs or air ducts;
established. This simple evaluation system could also be used as this caused fire and smoke to spread to higher floors. In
evaluation criteria for the government to promote the fire safety seven cases, there were 214 deaths and 1275 injuries, for
certification of existing hotel buildings. a disaster-cause ratio of C11 ¼ 0.77.
 Y.-Y. Chen et al. / Building and Environment 51 (2012) 311e319 313

Fig. 1. Fire safety evaluation system for existing multi-purpose hotel buildings.

(2) Lack of fire fighting equipment on emergency openings in 2. Disaster-cause dimension of evacuation safety:
exterior walls or proper distances between stations caused (1) Lack of an alarm system and delay in initiating evacuation
the fire to spread to higher floors. In two cases, there were resulted in 143 deaths and 693 injuries in six cases, for
105 deaths and 496 injuries, for a disaster-cause ratio of a disaster-cause ratio of C21 ¼ 0.539.
C12 ¼ 0.335. (2) The escape routes (Priority Safety Area) had decorations
(3) No fire compartment for guest rooms. This caused the fire consisting of flammable material that obstructed them,
to spread quickly, resulting in 12 deaths and 75 people resulting in 100 deaths and 196 injuries in two cases, for
injured in two cases, for a disaster-cause ratio of C13 ¼ 0.121. a disaster-cause ratio of C22 ¼ 0.411.
(4) No fire sprinklers or proper installation in four cases, which (3) The doors to the stairways were locked, making possible
led to 210 deaths and 767 injured, for a disaster-cause ratio evacuation unlikely. In two cases, there were five
of C14 ¼ 0.617. deaths and 56 injuries, for a disaster-cause ratio of
(5) Fire compartment was not complete in two cases, and C23 ¼ 0.084.
resulted in 28 deaths and 21 injuries, for a disaster-cause (4) Lack of an emergency power supply resulted in the failure
ratio of C15 ¼ 0.248. of the emergency light in two cases; there were no reported

Table 1
Fire incidents at multi-purpose hotel buildings from 1980 to 2006.

No Area Name of Hotel Incident time Floor location Number of Death Injury
guest rooms
1 Texas Hilton Hotel 1982 13F 306 12 e
2 Boston, MA Westin Hotel 1984 38F 840 0 46
3 Paterson, NJ Alexander Hamilton Hotel 1984 8F 180 13 50
4 San Juan, PR Dupont Plaza 1986 20F 289 97 146
5 Boston, MA Sheraton, Boston 1989 32F 2262 0 0
6 Las Vegas, NV MGM Grand 1980 23F 3000 85 600
7 Orlando, Florida Howard Johnson’s Hotel 1984 14F 242 0 75
8 Las Vegas, NV Las Vegas Hilton Hotel 1981 30F 2783 8 350
9 Kaohsiung, the ROC Hua Chi Hotel 1992 e 75 17 13
10 Taipei, the ROC The Grand Hotel 1995 14F 650 0 0
11 Chiayi, the ROC Festival Hotel, FortuneWorld 1995 14F 60 11 8
Building
12 Taoyuan, the ROC Four Season Guest House 2003 10F 80 5 10
13 Taitung, the ROC Hoya Resort Hotel, Zhibian 2004 e 315 0 12
314 Y.-Y. Chen et al. / Building and Environment 51 (2012) 311e319

deaths, but 46 people were injured, for a disaster-cause factors of safety was decided as the optimal evaluation of this
ratio of C24 ¼ 0.019. research [46].
(5) Hotel staff members were unfamiliar with emergency In the end, with the factor and require qualified factor of
response measures, so guests were guided in the wrong amended and modified strategies to achieve safety levels required
direction or used elevators (without smoke and fireproof by existing hotels, a matrix formula with a calculated evaluation
devices) during the evacuation. In six cases, there were 118 item weight was reviewed by each expert. The geometric average is
deaths and 706 injuries, for a disaster-cause ratio of used to integrate the matrix of each expert for the calculation of the
C25 ¼ 0.498. final weight.
(6) The doors to the fire escape stairs were not closed and From the evaluation results, 28 evaluation factors of safety
resulted in a stack effect which caused 219 deaths and 1239 (Attribute) were identified and in consideration of human safety, 13
injuries in seven cases, for a disaster-cause ratio of require qualified factors were listed, two from Strategy 1, four from
C26 ¼ 0.816. Strategy 2, and seven from Strategy 3. Please refer to factors with
(7) Inappropriate evacuation signs or functions in one case led the sign * in Table 2.
to 97 dead and 146 injured, for a disaster-cause ratio of Each strategy includes ([S i,j]k) safety evaluation factors (Attrib-
C27 ¼ 0.291. utei) and secondary evaluation factors (Itemj). The formula used to
(8) Insufficient fire and smoke compartments caused 231 calculate the evaluation factors of each strategy is shown as
deaths and 1288 injuries in eight cases, for a disaster-cause follows:
ratio of C28 ¼ 0.908.    
Si;j k ¼ Attributei ; Itemj k (2)
3. Disaster-cause dimension of location of fire:
(9) Fires which occurred in guest rooms led to 25 deaths and where k ¼ 1 e 3. k ¼ 1 is the fire prevention strategy, k ¼ 2 is the
50 injuries in three cases, for a disaster-cause ratio of evacuation and mitigation strategy, k ¼ 3 is the fire control and
C31 ¼ 0.134. resistance strategy, and k ¼ 1,2 represents the fire safety manage-
(10) Fires which occurred in an unused space in four cases led to ment measures. The Safety Evaluation Structure is shown in Fig. 2.
106 dead and 1075 injured, for a disaster-cause ratio of Through the Delphi Method, multicriteria decision-making
C32 ¼ 0.351. process was used in order to determine the weighting of each
(11) Fires which occurred at a spatial interface in two cases factor. We have applied The Analytical Hierarchy Process (AHP), and
resulted in 16 dead and 18 injured, for a disaster-cause ratio Analytic Hierarchy Process, AHP which was invented by Thomas L.
of C33 ¼ 0.032. Saaty in 1971 [47]. It is mainly used in an uncertainty and a decision
(12) Fires which occurred in the electrical control rooms or problem which has multiple evaluation criteria. This method has
converter rooms in two cases resulted in 58 people injured been used extensively in the level of building fire safety factors and
with no deaths reported, for a disaster-cause ratio of significance [3,48] so that the quantitative weighting can be
C34 ¼ 0.174. understood.
4. Disaster-cause dimension of fire causes: The evaluation method includes fire management factors.
(13) Inappropriate use of fire caused fire to spread in buildings People interacting is difficult to quantified and cognized. For
that did not have a designated fire compartment, resulting reducing the level variance [49], AHP is used to compare the
in 41 deaths and 68 injuries in four cases, for a disaster- significant level among factors so that evaluation weighting can be
cause ratio of C41 ¼ 0.309. carried out for factors.
(14) Wire or electrical fires or failure resulted in 90 deaths and We acquired [Si,j]1 ¼ 25.99%, [Si,j]2 ¼ 28.96%, [Si,j]3 ¼ 45.05%
656 injuries, for a disaster-cause ratio of C42 ¼ 0.163. (Table 2). While [Si,j]1, [Si,j]2 accounted for 54.95%, it was expected
(15) Arson normally occurring in unused spaces, idle lifts or that fire prevention management measures could be used as factors
safety stairs. In three cases, 105 people died and 571 were for evaluating the safety of existing hotel buildings and ensuring
injured, for a disaster-caused ratio of C43 ¼ 0.442. their basic safety. The experts suggested that the total score of the
(16) Incompetent construction methods led to one case where, qualified factors should account for 1/2 of the total distribution and
due to good response, no casualties were reported. must include 13 items: [S2,1e6]1, [S5,1e5]1, [S2,1e4]2, [S3,1e4]2, [S4,1e3]2,
[S8,1e2]2, [S2,1e4]3, [S3,1e4]3, [S4,1e5]3, [S7,1e2]3, [S8,1e6]3, [S9,1e3]3, and
According to the four causal dimensions and the disaster-cause [S10,1e4]3 (* indicated in Table 2).
level, poor fire management led to a high level of disaster-caused We tried to use the matrix at 9  6 to show each item of weight
fires. Therefore, having taken into consideration the difficulty of in order to evaluate the existed building. When k ¼ 1, [Si,j]1, the
upgrading the hardware of existing buildings, this study based weight is shown in the following matrix:
its fire management scheme on three safety strategies: fire 2 3
1:0161 0:7621 0:8467 0:8467 0:7621 0
prevention, fire evacuation and mitigation, and fire control and 6 0:3387 0:5080 0:8467 1:0161 0:3387 0:3387 7
resistance. The fire management scheme incorporated these strat- 6 7
6 2:3709 2:7096 0 0 0 0 7
6 7
egies into daily maintenance procedures, evacuation drills, evacu- 6 0:4234 0:6774 0:6774 0 0 0 7
ation and fire mitigation strategies, as well as equipment or facilities.   6 7
Si;j 1 ¼ 6
6 1:0161 0:6774 1:0161 0:6774 0:2540 0 7 7
To fit the characteristics of existing hotels, this study uses the 6 1:3548 0 0 0 0 0 7
6 7
Delphi Method, a good and effective one for group discussion, to 6 1:0161 0:8467 0:8467 0 0 0 7
6 7
acquire evaluation factors, key factors (require qualified factors), 4 1:6935 0 0 0 0 0 5
and their importance [44]. Researchers first acquired an initial 0:6774 0:6774 1:0161 0 0 0
evaluation factor of disaster-cause ratios from literature review and
the above four dimensions [45] and then analyzed questionnaire
responses from 50 experts (in the fields of building and fire 3.2. Potential risks of use behaviors in existing multi-purpose hotel
designer, insurance risk evaluation engineer, and hotel manager) to buildings
determine evaluation factors and their importance.
A second round meeting was conducted and based on the The high-floor layout and multi-purpose use of hotel buildings
consideration of human safety and evacuation, the require qualified increase their risk hence, Potential risks involve multiple spatial
 Y.-Y. Chen et al. / Building and Environment 51 (2012) 311e319 315

Table 2
Fire safety evaluation factors of existing multi-purpose hotel buildings.

Strategy k ¼ 1 Strategy k ¼ 2 Strategy k ¼ 3

Fire prevention (25.99%) Evacuation and mitigation (28.96%) Fire control and resistance (45.05%)
i ¼ 1. Duties of fire management organization system i ¼ 1.Utilization of emergency broadcasting equipment i ¼ 1. Fire resistance performance of building
(4.2%) (4.2%) structure (2.5%)
j ¼ 1e5 Fire prevention awareness of managers, duty j ¼ 1e4 Installation, links, and contents of broadcasting j ¼ 1 Fire resistance of pillars, columns, walls,
staff, and organization. equipment floors, and roof
i ¼ 2. Fire safety plans and self organized drills (3.4%)* i ¼ 2. Evacuation facilities (4.1%)* i ¼ 2. Material of interior decoration (3.4%)*
j ¼ 1e6 Contents of management plan, drill of j ¼ 1e5 Installation and feasibility of safety area, j ¼ 1e4 Flame resistance of location
emergency procedure and drill evacuation platform
i ¼ 3. Fire prevention awareness (5.1%) i ¼ 3. Fire escape equipment (3.3%)* i ¼ 3. Horizontal fire compartment (4.2%)*
j ¼ 1e2 Task and capabilities and awareness of fire, j ¼ 1e4 Installation and feasibility of emergency j ¼ 1e4 Safety area and fire resistance of
and duties of fire-fighting personnel lighting, signage, and evacuation devices, escape sling compartment
and drill
i ¼ 4. General equipment management and i ¼ 4. Utilization of evacuation facilities (2.7%)* i ¼ 4. Fire barriers with vertical openings (4.2%)*
maintenance (1.8%) j ¼ 1e3 Accessibility and recognition of entrance and j ¼ 1e5 Vertical fire planning of ducts,
j ¼ 1e3 Effectiveness of maintenance of electrical, air exit, evacuation route elevated areas, stairs, elevators, and
conditioning, and gas equipment openings of exterior walls
i ¼ 5. Maintenance of fire equipments and facilities (3.4%)* i ¼ 5. Utilization of fire escape equipment (1.9%) i ¼ 5. Fire door (8.5%)
j ¼ 1e5 Daily check on fire fighting equipment, fire door, j ¼ 1e3 Suitability and effectiveness of evacuation j ¼ 1e3 Accessibility and fire- and smoke-
and escape routes equipment and signage proof function of fire doors, controlled links
between fire doors and fire prevention rolling
doors
i ¼ 6. Security system or monitoring system (1.4%) i ¼ 6. Fire fighting capacity (2.0%) i ¼ 6. Spatial characteristics (4.9%)
j ¼ 1 Monitoring system j ¼ 1e3 Fire fighting support, such as distance to fire j ¼ 1e3 Type of fire and smoke spread of
fighting brigade and building height evacuation routes
i ¼ 7. Fire and gas equipment (2.7%) i ¼ 7. Mandatory fire fighting measures (4.7%) i ¼ 7. Automatic alarm equipment (4.2%)*
j ¼ 1e3 Environmental conditions of kitchen and furnace j ¼ 1e5 Condition and installation of peripheral alleys, j ¼ 1e2 Installation and utilization of fire
rooms emergency elevators, and rescue space detectors and alarm systems
i ¼ 8. Performance of power supply equipment (1.7%) i ¼ 8. Safety of openings of exterior walls (3.4%)* i ¼ 8. Fire extinguishing equipment (5.4%)*
j ¼ 1 Generation function of power generation system j ¼ 1e2 Installation and accessibility of emergency j ¼ 1e6 Fire extinguishing equipment, such as
entrance and exit fire extinguisher, fire hydrant and automatic
sprinkler.
i ¼ 9. Report status of fire safety check of building i ¼ 9. Risk transfer measure (2.5%) i ¼ 9.Smoke exhaust equipment (3.4%)*
facilities and fire equipments (2.4%) j ¼ 1e2 Insurance and fire incident history j ¼ 1e3 Planning of smoke prevention area and
j ¼ 1e3 Check and realization of building safety and smoke exhaust methods
hardware check of fire fighting safety equipment
i ¼ 10. Utilization of automatic alarm system
and alarm bell (4.2%)*
j ¼ 1e4 Links and report transfer system of
automatic alarm equipment
*
Qualified factors.

uses etc. In order to evaluate potential risk factors, five experts 1. Type of Business:
engaged in hotel construction, fire design and hotel management (1) There are many multi-purpose use characteristics and fire
were interviewed to identify five risk characteristics related to use risk factors in buildings. The spatial utilization on the higher
behavior. floors of multi-purpose hotel buildings usually includes
After reviewing the fire protection and fire safety policies of restaurants and kitchens, with meeting rooms and shops on
hotel buildings with high level floors in Taiwan, the researchers the lower floors. Fire risks consist of smoke filter masks
understood the risk characteristics to be as follows: (pipes), heaters, gas, electrical sources and cigarette butts.

Fig. 2. Safety evaluation structure.

316 Y.-Y. Chen et al. / Building and Environment 51 (2012) 311e319

(2) Under multi-purpose use, an open restaurant is used as a secondary consideration, although safety equipment (facilities)
a kitchen and restaurant, or shopping malls are used as belongs to depreciated hardware and maintenance and updates are
meeting rooms or restaurants, which create links and required to maintain function.
interactive interfaces. Firefighters find it difficult to control
the links between fire doors (fire rolling shutter) and (2) Frequent interior decoration and changes.
detectors, and there is an inter-connection of evacuation
signage and buildings, and security requirements. In order to attract consumers and rentals, restaurants and shops
2. Short-term occupants: frequently change their decor. In order to provide innovative
services, re-modeling is required for guest rooms every five years
Hotels accommodate non-specific people of various ages and on average.
cultures who generally have a low level of environmental aware-
ness and who need to be guided during evacuation procedures to 5. Fire prevention and training issues:
reduce casualties. (1) High turnover rate of operating staff.

3. Fire management plans: Operating staff normally has a cooperative relationship with
(1) Characteristics of dual focus on security and safety. hotels (or restaurants), so turnover rate is high. Awareness at the
managerial level dominates whether or not fire prevention
Users of hotels demonstrate short-term consumption behaviors education is given to new employees.
(dining and shopping). They are short-term occupants (accommo-
dation) who desire privacy and security. If the safety ladders or (2) Focus levels vary according to the fire awareness of
elevators on the consumption floors of the buildings are locked, managers.
accessibility to the evacuation routes is obstructed. (3) Under multi-purpose use behaviors, there generally is no
overall fire prevention and emergency response training.
(2) Characteristics of big management organization.
Parameters defined according to risk characteristics of hotel
Engineering and safety departments involved in fire prevention buildings are the height of buildings, multiple-purpose use
safety are administered by either management or procurement behaviors, occupancy rate and the allocation of repair and main-
departments. Due to different management organizations and tenance expenses. A five-parts is used to calculate risks (the highest
duties, resources available to these two departments vary. is evaluated with positive scores) and a reduced value of potential
use risks (URred) is decided by experts.
(3) Difficulties managing fire prevention safety are due to Due to potential user risks in hotel buildings, the four risk
a greater focus on dining services. parameters were determined to be: building height, multi-purpose
use behaviors, occupancy rate of guest rooms and maintenance fee
According to the 2010 Tourism Hotel Operation Monthly Report allocation to acquire a reduced value of potential use risks, URred:
released by the Tourism Bureau, the Ministry of Transportation and 8 9
Communications, the Republic of China, dining revenues contrib- >
> 0:8 ; r0 ¼8 >
>
< =
0:9 ; r0 ¼ 0w7 0
uted to 45% of the total turnover of hotels. Due to the focus on URred ¼ Where r ¼ B þ C þ L þ M (3)
>
> 1 ; r0 ¼ 4w0 >
>
dining services, daily maintenance and management were difficult : ;
1:2 ; r0 ¼ 6w  4
to maintain and most evacuation routes were occupied so fire doors
in kitchens were dismantled. The r0 is the reduced value that was determined by the four risk
parameters (Table 3).
4. Enhancement of service facilities:
(1) The concept of costly enhancements for building safety. 3.3. Fire safety evaluation grading system

It is difficult to calculate the actual income resulting from the In order to understand whether the fire safety of a hotel building
enhancement of facilities. Therefore, facility enhancement is was acceptable or not, improvement proposals were determined

Table 3
Reduced value of potential risk parameters.

Risk parameter Reduced value r0

a
Building Height (B) S50 m 25 me49 m 21e24 m
þ2 þ1 0
Multi-Purpose Use More than 1/2 1/3e1/2 1/3e1/4 Less than 1/4 Restaurants Only Used
Behaviors (C)b Area of Scale Area of Scale Area of Scale Area of Scale by Hotel Guests
þ2 þ1 0 1 2
Occupancy Rate (L)c S70% 50e69% 49e30% &29%
þ2 þ1 0 1
Maintenance Fee <1.14% 2.18%e1.14% 2.18% 2.18%e6.9% More than 7%
Allocation (M)d þ2 þ1 0 1 2
a
There is a positive correlation between the number of floors in a building and fire loss. According to statistics on fires on the higher floors of hotels [21] and NFPA101 [10],
floors 4, 7 and 13 are the dividing points (each floor is calculated at 3 m). In addition, this study referred to L.T Wong, W.K Chow [30,50] and listed high-floor buildings as those
higher than 24 m.
b
There is a positive correlation between the area scale of restaurants to a total hotel area and danger risk.
c
Occupants of hotels are there for the short-term and have low environmental awareness. A high occupancy rate results in higher casualties during disasters.
d
Buildings and facilities after construction depreciate with time as do their functions. Equipment in buildings should be maintained and functions should be enhanced to
boost safety. Therefore, the repair and maintenance rate should have a negative correlation with danger. This study referred to the 2007 Operational Analysis Report of
Tourism Hotels in Taiwan released by the Tourism Bureau, Ministry of Transportation and Communications, the ROC [29].
 Y.-Y. Chen et al. / Building and Environment 51 (2012) 311e319 317

Table 4 priority improvements and items with low scores in Strategy 3

Evaluation grade criteria. were proposed to improve safety levels.
Level (R) VP P M S VS 2. The evaluation of Case 7 also found an ‘immediate danger’.
(Immediate (dangerous) (acceptable ) (safe) (very Strategies 1 to 3 evaluated this case as ‘dangerous’ and it was
danger) safe) determined that the qualified factors did not meet the
% of Scores &30% 31%e50% 51%e60% 61%e75% S76% requirements, indicating the building’s lack of basic safety and
poor fire management measures. Thus, the qualified factors of
Strategies 1 to 3 were listed for priority improvement. First, the
with fire fighting professionals to improve the fire safety of the qualified factors for Strategies 1 and 2 would be improved to
building. As a result, for this study, each evaluation indicator was meet the requirements; then the items of Strategy 3 with lower
calculated and classified according to the following five evaluation scores would be improved for the building to attain a safe level.
grades: VP (immediate danger), P (dangerous), M (accepted), S 3. The evaluation of Case 11 determined the building to be in
(safe) and VS (very safe). Each level was classified according to ‘immediate danger’. Although Strategy 1 evaluated this case as
a percentage of the total score as: safe, Strategies 2 and 3 evaluated it as dangerous and the

  
R ; R ; R ; R when qualified factors reach 1=2 of total scores
R ¼ Min  1 2 3 g  (4)
R1 ; R2 ; R3 ; Rg  1 when qualified factors fail to reach 1=2 of total scores

where R1eR3 are the corresponding grades of reduced scores

calculated with fire safety strategies, Rg is the general grade of
reduced scores after the calculation of all indicators; R is the lowest
grade of R1, R2, R3, Rg and when qualified factors do not account for
1/2 of total distributed scores. The evaluation grade criteria are
shown in Table 4.

4. Empirical evidence of cases for evaluation

For this paper, 16 cases of existing multi-purpose hotel buildings

were evaluated and it was found that five cases needed to be
improved to an acceptable level (4 VPs and 1 P) and 11 had
acceptable levels or better (9 M and 2S) (Table 5).
According to the strategic evaluation and distribution of each
case study (as shown in Fig. 3), five of the cases, #s 6,7,10,11 and 16,
had unacceptable levels of danger or were in immediate danger. As
a result, improvement plans were proposed to enhance the fire
safety of the buildings.

1. The final evaluation level result for Cases 6 and 10 was one of
‘immediate danger’ and, in addition to the evaluation of danger
in Strategy 3, the qualified factors did not meet the require-
ments. The buildings in these two cases were not sufficiently Fig. 3. Safety evaluation distribution of 16 Case studies.
safe. Thus, the qualified factors of Strategy 3 were listed as

Table 5
Empirical cases of existing multi-purpose hotel buildings.

Case No Number of hotel Maximum occupants Total floor area Building URred R1 R2 R3 Rg Qualified R
rooms (people) (m2) floor factors
1 856 207,357 118,805.5 26a 0.9 S S S S Reached S
2 606 137,548 52,696.65 14a 0.9 M S M M Reached M
3 420 104,018 79,467.14 43a 0.9 M S M M Reached M
4 343 94,361 43,671.36 20a 0.9 S S M M Reached M
5 336 51,673 87,292.93 38 0.9 VS S M S Reached M
6 276 47,804 46,644.61 13a 0.9 M S P M Failed to reach VP
7 245 50,600 93,281.84 17a 0.9 P P P P Failed to reach VP
8 220 55,303 13,060.047 11a 0.9 S S M S Reached M
9 209 58,688 16,337.82 12a 0.9 S S M S Reached M
10 208 49,419 23,762.35 22a 0.9 M S P M Failed to reach VP
11 200 1513 37,342.19 12 0.9 S P P M Failed to reach VP
12 200 64,361 21,284.21 12 0.9 S S M S Reached M
13 197 45,322 37,342.19 12 0.9 S S M M Reached M
14 183 36,214 15,171.09 7a 0.9 S S S S Reached S
15 112 20,399 8280.6 7a 0.9 S M M M Reached M
16 198 46,400 18,912.96 9 0.9 S P P M Reached P
a
Buildings over ten years old.
318 Y.-Y. Chen et al. / Building and Environment 51 (2012) 311e319

qualified factors did not meet the requirements, thus indicating (2) Ensure the safety and emergency facilities in the buildings:
the building’s lack of basic safety. Therefore, the qualified
factors and items with the lowest scores of Strategies 2 and 3 Flammable material shall be avoid to decoration in the evacu-
were proposed to improve the safety level. ation path. A sufficient number of smoke detector shall be installed
4. Case 16 was evaluated as ‘dangerous’. Although with Strategy 1, to increase evacuation time needed. It is also necessary to design
this case was evaluated as safe, with Strategies 2 and 3 it was evacuation signs properly with spatial layout of buildings.
evaluated as dangerous. Therefore, the qualified factors and This study attempts to use fire management measures to
items with the lowest scores of Strategies 2 and 3 were improve the safety of existing hotel buildings. However, basic safety
proposed to improve the safety level. function is still required by hardware facilities in the building. Such
5. The empirical cases determined there were five buildings in as suitable of fire safety zone, evacuation stair and use of fire
‘danger’ or in ‘immediate danger’. Of these, three were resistant decorative materials. In addition, skillful Interior design
buildings which had been established more than ten years shall also be used to improve the planning of a fire protection zone
prior, and two cases had failed to implement a fire safety as well as to protect the existing structure and appearance of hotel
management scheme and were evaluated as dangerous buildings. Thus, only with the cooperation of fire engineers, hotel
with Strategies 1 and 2. The results showed a lack of fire safety managers, and interior designers can the proposal for optimal
management for old buildings. improvement be compiled with.
This research result has been incorporated into Fire Safety Logo
Certification promoted by the Taiwanese government and the
5. Conclusion research time is short. Various issues shall be left for future
examination (such as the relationship between and among Strate-
This study proposes a safety evaluation method based on the gies 1, 2, and 3 as well as the co-relationship). However, this set of
risk of human behavior of hotels [18]. However, it cannot be methods provide help to improve the safety level of existing
applied to all types of buildings. The evaluation method attempts buildings without major reconstruction as well as solve the diffi-
to improve insufficient safety of existing multi-purpose use hotel culties of walking distance and the required number of entrances
buildings with fire safety management. Thus, three fire safety perceived in literature review [1,3,4,12,14,15].
strategies are proposed and evaluation risks are acquired from
a questionnaire investigation of 50 experts with Delphi Method.
The AHP method is used to acquire a factor weight including fire Acknowledgments
prevention strategy, evacuation and mitigation strategy for 54.95%,
fire control and resistance strategy for 45.05%. This study also During the research time, the research team received a kind
proposed the concept of a require qualified factor to meet the subsidiary budget from the National Science Council (Project
basic safety level of existing hotel buildings. Number 957-01-6489-1). It was truly their unreserved assistance
This study also proves that among the evaluations of the multi- that helped in accomplishing this research smoothly, and to them
purpose use hotel buildings of 16 cases. The five of them are under this research shall always be obliged.
the minimum safety level. After reviewing each strategic factor, an The author would like to thank ABRI (Architecture and Building
improvement proposal is submitted for the certification of the Fire Research Institute, MOI) and TABC (Taiwan Architecture and Buil-
Safety Logo to protect the safety of millions of tourists in Taiwan. dingCenter) which provide the chance to improved on Fire Safety
The empirical evaluation found that: Logo certificate.

1. In the empirical study, 5/16 was evaluated as the last rank (R)
References
bellow safety level and fire control and resistance strategy is
also below the safety level, showing a positive co-relation [1] Chow WK. Proposed fire safety ranking system EB-FSRS for existing high-rise
between R and R3. Nonresidential buildings in Hong Kong. J Architect Eng 2002;8(4):116e24.
[2] SFPE handbook of fire protection engineering. 3rd ed. 2002.
2. In the empirical case study, 12/16 require qualified factors
[3] Lo SM. A fire safety assessment system for existing buildings. Fire Tech 1999;
scored higher than required level indicating most hotel build- 35(2):131e52.
ings in Taiwan meet the basic safety level but 4/16 fail to meet [4] Liu M, Lo SM. On the use of fuzzy synthetic evaluation and optimal classifi-
the safety requirement due to damaged fire compartment and cation for computing fire risk ranking of buildings. Neural Comput Appl 2009;
18(6):643e52.
insufficiently flame resistant decorative material. [5] Rasbash D, Ramachandran G, Kandola B. Evaluation of fire safety. John Wiley
3. Factors of Strategy 3 are found with insufficient fire facilities or and Sons; 2004. 366.
equipments. Sometimes, improve the Strategy 1 and 2 factors [6] Gretener M. Evaluation of fire Hazard and determining Protective measures.
Association of Cantonal Institutions for Fire Insurance and Fire Prevention
will be ensured the performance of fire facilities and Service for Industry and Trade Zurich; 1973.
equipments. [7] Guide on alternative approaches to life safety, NFPA-101A, National Fire
4. The require qualified factors can conservatively be used as the Protection. NFPA; 1995b.
[8] Watts Jr John M, Kaplan Marilyn E. Fire risk index for historic buildings. Fire
final watchdog and to remind building owners to self evaluate Tech 2001;37(2):167e82.
the safety level of their buildings. [9] Watts Jr John M. Analysis of the NFPA fire safety evaluation system for busi-
5. According to the empirical results, the following fire manage- ness occupancies. Fire Tech 1997;33(3):276e82.
[10] Guide on alternative approaches to life safety, NFPA 101A; 2007 [Edition].
ment measures as should be adopted to improve safety:
[11] Chow WK, Wong LT, Kwan ECY. Proposed fire safety ranking system for old
(1) Measures to improve fire safety management: high-rise buildings in the Hong Kong special administrative region. Fire Mater
1999;23(1):27e31.
[12] Zhao CM, Lo SM, Lu JA, Fang Z. A simulation approach for ranking of fire safety
Usability and signal of fire detectors, fire alarms and fire
attributes of existing buildings. Fire Saf J 2004;39(7):557e79.
pumps. Evacuation equipment shall be enhanced; stationary [13] First draft of interim technical guidelines on the inspection, assessment and
opened doors with sensors shall be installed. Owners shall repair of buildings for the Building Safety Inspection Scheme(BSIS). Building
voluntarily maintain and inspect fire protection equipment. Users’ Department, Hong Kong Government; 1997.
[14] Lo SM, Hu BQ, Liu M, Yuen KK. On the use of reliability interval method and
behavior of buildings shall be incorporated into contingency grey relational model for fire safety ranking of existing buildings. Fire Tech
planning. 2005;41(4):255e70.
 Y.-Y. Chen et al. / Building and Environment 51 (2012) 311e319 319

[15] Wong LT, Lau SW. A fire safety evaluation system for prioritizing fire [32] Code of inspection and reporting of fire safety equipments on use and occu-
improvements in old high-rise buildings in Hong Kong. Fire Tech 2007;43(3): pancy. Taiwan: National Fire Agency; 2008.
233e49. [33] Code of inspection and reporting of fire safety facility. National Building
[16] Chow WK, Lui GCH. A fire safety ranking system for karaoke establishments in Agency; 2010.
Hong Kong. J Fire Sci 2001;19(2):106e20. [34] Code of improvement of fire safety facility and fire safety equipments in the
[17] Chow WK, Lui GCH. A proposed fire safety ranking system for karaoke existing buildings. National Building Agency; 2007.
establishments and its comparison with the NFPA-fire safety evaluation [35] Richard Best, David P. Demers, P.e., 1980.11 MGM Grand, Las Vegas, NV, NFPA
system. Build Environ 2002;37(6):647e56. Fire Investigations report.
[18] Kobes M., Fire risk of high-rise buildings based on human behavior in fires”, [36] David P. Demers, P.E., 1981.2Hilton, Las Vegas, NV, NFPA fire investigations
Conference Proceedings FSHB 2008. First International Conference on fire report.
Safety of High-rise Buildings. Bucharest, Romania, May 07e09, 2008. [37] Pamela Powell, 1982.3.6 Hilton Hotel-Houston TX, NFPA fire investigations
[19] Kobes M. Way finding during fire evacuation; an analysis of unannounced fire report.
drills in a hotel at night. Build Environ 2010;45(3):537e48. [38] Tom Timoney, 1984.1.28Howard Johnson, Orlando, FL, NFPA fire investiga-
[20] Jennifer Flynn D. U.S. HOTEL and MOTEL STRUCTURE FIRES. National Fire tions report.
Protection Association; 2010. NFPA Report. [39] Rita Fahy, Tom Timoney, 1984.1.2 Westin Hotel, Boston, MA, NFPA fire
[21] John R. Hall, Jr., HIGH-RISE BUILDING FIRES”, NFPA Report, 2009. investigations report.
[22] Update of response time loss relationships for the fire service emergency [40] Carl E. Peterson, 1984.10.18 Alexander Hamilton Hotel, Paterson, NJ, NFPA fire
cover toolkit e fire research report 3/2010. UK: Greenstreet Berman Ltd., Investigations report.
Department for Communities and Local Government; 2010/12. [41] Thomas J. Klem, 1986.12.31 Dupont Plaza, San Juan, PR, NFPA Fire Investiga-
[23] FiRE-TECH, Fire risk assessment methods e European study into the fire risk tions Report.
to European Cultural Heritage. European Commission; 2003. 8. [42] Michael S. Isner, 1989.8.30 Sheraton, Boston, MA, NFPA fire investigations
[24] Kobes M. Zelfredzaamheid bij Brand- Kritische Factoren voor het Veilig Report.
Vluchten uit Gebouwen”. Boom Juridische Uitgevers, Den Haag. (in Dutch), [43] National Fire Agency, Ministry Of The Interior, http://www.nfa.gov.tw.
ISBN 978-90-5454-850-8; 2008. [44] Hsu Chia-Chien. The Delphi technique: making Sense of Consensus, practical
[25] Day K. Staying safe. Fire Risk Manag; August 2009:48e50. assessment. Res Eval 2007;12(10).
[26] Demers P. Hotel fire. Las Vegas, NV; February 10, 1981. Quincy, MA: NFPA; [45] Okoli Chitu, Pawlowski Suzanne D. The Delphi method as a research tool: an
1981. example, design considerations and applications. Inform Manag 2004;42(1):
[27] Bryan JL. Human behavior and fire. In: NFPA handbook, Section 7, Chapter 1. 15e29.
Quincy: NFPA; 1992. [46] Kerlinger FN. Foundations of behavioral research. New York: Holt, Rinehart,
[28] Richardson LR. Fire losses in selected property classifications of non- and Winston, Inc.; 1973.
residential, commercial and residential wood buildings. Part 1: hotels/ [47] Saaty TL. The analytic hierarchy process. New York: McGraw-Hill; 1980.
motels and care homes for aged. Fire Mater 2007;31(2):97e123. [48] Zhao CM, Lo SM. A simulation approach for ranking of fire safety attributes of
[29] Yearly statistics reports of Taiwan hotel rooms and monthly statistics reports existing buildings. Fire Saf J 2004;39(7):557e79.
of operational statistics. Taiwan: Tourism Bureau; 2006e2010. [49] Saaty TL. How to make a decision: the analytic hierarchy process. Eur J Oper
[30] Chow WK. Building fire safety in the far East. Architect Sci Rev 2005;48(4): Res 1990;48(1):9e26.
285e94. [50] Wong LT, Chow WK, Kwan Eric CY. A brief review on fire regulations for old
[31] Chief Executive, Marguerite Seqauaris, Guidelines to fire safety in European highrise commercial buildings in Hong kong. Int J Eng Perform-Based Fire
hotels. HOTREC; 2010. Codes 2000;2(4):153e60.