See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/316619782

A Case Study on Investigation of Fire Accident Analysis in Cotton Mills

Conference Paper · December 2016

CITATIONS READS
0 8,914

2 authors, including:

Rajpradeesh Thangaraj
Kalasalingam University
3 PUBLICATIONS   1 CITATION   

SEE PROFILE

All content following this page was uploaded by Rajpradeesh Thangaraj on 02 May 2017.

The user has requested enhancement of the downloaded file.

 14th International Conference on Humanizing Work and Work Environment HWWE-2016
ISBN Number: 978-93-83006-81-6

A Case Study on Investigation of Fire Accident Analysis in Cotton

Mills
Shanmuga Sundaram P, Raj Pradeesh. T, S. Aravind Raj and Jaya krishna. K

Abstract—Textile industry is main production sector to The main common reasons of fire accident in any place
improve economy of country. It consists of processing units like either knocking liquid on electrical equipment and smoking
cotton mill, spinning mill, dyeing units, garments exports. in work place, handling hazardous material or chemical with
Cotton industry primarily used to convert raw cotton into yarn have proper knowledge are all the foreside example may
product. Fire is a main hazard present in all process area of become the major causes for fire accidents.
production field. The main causes of fire have many reason that
static electricity, overheating, human error, environmental
II. LITERATURE REVIEW
atmosphere and failure of electrical parts in machine. As per
statistics date given from National Crime Record Bureau Evacuation timing has valuable factor while calculation of
(NCRB), the number of person died due to stumpage is higher fire accident prevention plan. The distance between the fires
than people affected by fire at the time of fire accident. The occur location and exit of building should be take account of
samples are evaluated with hypothesis and T-Test. The position consideration in fire safety point of view. The evacuation
of fire emergency door is decided based on the result of t-test timing has been calculated through simulation models and
and evacuation timing calculation. mathematical formulas of Monte Carlo simulation format
[1]. The article has discussed the new method of calculating
I. INTRODUCTION theoretical method of evacuation with Pathfinder simulation
The process of textile industry contains several parts like software. The theoretical calculation can not only calculate
cotton mill, spinning mill, weaving unit and garments. Every the total time of people evacuation necessary time but also
manufacturing unit and residential area has possible of consider people retention time period in different width of
accidents and hazards. The major hazards are known as place and specified fire safety design [2].
noise, fire, dust, temperature, ventilation, slippery. The The concept of this article gives that to minimize the time
objective of work is to concentrate on the basic of fire taken for evacuation should be minimized in heuristic
accidents happened in various industries and cotton mills, algorithm model. Evacuation path group and evacuation
textile field. The data are collected from various sources like timing and number of evacuation is ready to use are
reference site and research articles contents. The possible of connected with evacuation network constantly to find correct
accidents has increased due to improper maintenance and plan [3]. Behavioral of person in work place acts as major
human mistake. impact role of evacuation time to reach safe place. The
This study was carried out in cotton mill located in article has collection of behavioral concepts helps to arrange
Ichekaranji, Maharashtra, India. The number fire accidents evacuation decision model [4].The part of article has been
are very high as compared to other state in India. The data explained about the inherent safety system and safety
has collected from National Crime Records Bureau (NCRB). equipment and protection level in company. The remaining
The electrical faults are the prime cause for work place part of research work has been contained emergency action
fire and loose wires, plug on over loaded and old equipments plan and intervention timing and implement of the action in
are can make for probable death trip. All work area should right situation [5]. Fault tree analysis has been used to find
be monitor with Portable Appliance Testing (POT). The test out the main factors for fire accidents in high rise building.
should be taken some reasonable period of time interval. Evacuation of all people from the building is not easy task at
The flammable and combustible materials also act main fire accidents, the problem has discussed with elevator and
factor for industrial fire accidents. The storage and stair way model. The elevator evacuation plan is suitable for
transportation of hazardous material must be following the old and sick people, children from the high rise building to
standards of explosive act 2003(division4, clause 84). safe area [6]. The physical and chemical properties of cotton
Controls of material flow have to analysis at the time of risk have discussed on part of article and the possible factors are
assessment. listed out based on event and fault tree analysis. The utility
of space has major aspect for fire accident and spread of fire
into work area. The capacity of water storage and
firefighting equipment has to arrange before establishment
F. Shanmugasundaram P, Post Graduate student, Department of of any cotton logistics ware house [7].
Mechanical Engineering -Kalasalingam University, Srivilliputtur, Activity theory has discussed in case study based on
Tamilnadu, India(e-mail:sundaram.ise@gmail.com) human related accidents. The theory gives holistic
S. Raj Pradeesh T, Aravind Raj S, Department of Mechanical contextual back grounds for understanding and diagnosing
Engineering, Kalasalingam University, Srivilliputtur, Tamilnadu, India (e-
mail: t.rajpradeesh@klu.ac.in) human related accidents. Human Factor Analysis &
T. Jayakrishna K , School of Mechanical Engineering, VIT University, Classification system (HFACS) was also applied to that case
Vellore, India

236
 14th International Conference on Humanizing Work and Work Environment HWWE-2016
ISBN Number: 978-93-83006-81-6
study. The method will be use when the need of human 2013 23,593
activity factor derived from strong theoretical model [8]. 2014 20,377
Fig 1: Accidents figures data in 2010-2014.
III. HISTORY OF MAJOR FIRE ACCIDENTS IN INDIA According to NCRB data the fire accidents had
AND BANGALADESH filled in different location of India. Among the total number
of accidents was highlighted in 2011 as 26,343. The main
Bangladesh is an important country for export textile and reasons for the accidents were analyzed by special teams.
garment materials into European and German countries. As They have suggested that improper safety policy and
like the countries capital Dhaka is famous for fire accidents working condition must be corrected within the specified
in garments and textile industries. Each and every year the time interval [14].
country had met at least more than two fire accidents and
building collapse in business and residential area [9]. B. REASON FOR DEATH DUE TO FIRE
On 25 Nov 2012 the worst fire accidents were occurs in
capital of Bangladesh in eight stores Garments Company. Suffocation
More than 125 people were died due to fire and stumpage. 8 12
The local residents people were reported this issue to fire 4 Lack of
Ventilation
fighters and police. They had seen that some of the workers 25
had jumped from the windows of eighth to prevent their life. 17 Locked Doors
The main door of company was closed always to prevent
from thief and non authority persons entry into company at Blocked Exit
13 Route
the time of working hours [10]. 21
Rana Building collapse was very worst accident in Smoke
Bangladesh. More than 1132 garments workers were killed
in that accident on 23 April 2013. From this accident, the Fig 2: Reason for Death due to fire accident
county had arranged special panel with seven members. The According to statics data collation, the main causes of
panel had advised to all garments and textile mills, export death at the time of fire accidents occurs in any place which
companies about safety of work process and safe is not only textile and chemical process industries but also
environment of workers [11]. residential area such as shopping mall, marriage hall. The
The garment and textile factory fire accidents since maximum numbers of people or workers died due to
Tazreen Fashion Factory fire accidents 24 Nov 2012 up to stampede in that place. The stampede gets 25% of priority as
2015 were documented by the solidarity center of highest level in fire accident. Blocked exit doors and locked
Bangladesh. The document has been used to identify the doors were related reason for the stampede in accident area.
main reason of accidents and the details of injury people and The direct effect of fire at the time of accident was only
fatal workers counts. The total number of fire accident and possible of 8% to the workers or people who presented very
building collapse were filed as 89 accidents on 2012-2015. near to source of fire location. The remaining reasons are
The total number of people were injured 903 people closely connected with the behavior of people and unsafe
including workers and nearby resident people [12]. condition of work place in factory or residential area. [15]
According to survey of Times of India at April 11, 2016,
over 1.13 lashes of people killed in fire accidents in past 5 IV. METHODOLOGY
years, the numbers of accidents were high in Maharashtra
state. The death over 100 people in Kerala temple has Safety review is a tool of hazard identification in any work
retained the focus on fire related accidents. Data documents place. The review contains the interaction with workers and
were released by the National Crime Records Bureau supervisor with basic safety questions. The response of
(NCRB) of home ministry; it had showed that 20,377 fire workers related to safety aspect helps to find out the most
accidents were reported in our country in 2014 which caused hazardous area in work place.
19.513 deaths and injuries to 1889 people [13].
The maximum number of dead people has been figured
from Maharashtra state (3892), followed by Madhyapradesh
(2305), Gujarat (2011), Karnataka (1610), Tamilnadu
(1594). Other cases includes fire in school building , mines ,
trains , private vehicles, Government buildings,
manufacturing units of combust able material like fire
crackers and chemical , match boxes [13].

A. ACCIDENT DATA BY NCRB

Year Accident
2010 26,025
2011 26,343 Fig 2: Methodology
2012 24,987

237
 14th International Conference on Humanizing Work and Work Environment HWWE-2016
ISBN Number: 978-93-83006-81-6
V. CASE STUDY AND COMPUTATION
Accounting to result of review, the speed room section
has higher possible of fire contaminants. The location of
speed room department is very near to cotton quality
checking lab and workers rest room. Speed room
departments have equipments with very high rotation speed
nearly 15000 RPM. It is an only department has higher range
of speed machine in company. Observation of workers
behavior in company may potential for fire due to reason of
smoking habit in rest room [16].
The machine has been working continuously for 24 hours
shift. Failures of electrical equipments are kept in corner of
Fig. 3 Sample position of fire and emergency door
that room. Level of heat has been rise in surrounding of any
equipment as a result of high rotation and long time.
Above diagram represent the position of fire and worker
A. Mock fire drill:
running path, location of emergency door placed as
A fire drill is a method of practicing how a building would imaginary. The priority has given by workers to reach the
be evacuated in the event of a fire or other emergency. normal and emergency door. The evacuation running paths
Active fire protection system such as fire alarm is activated
are clearly noticed. There were 3 different way of approach
and the people or workers will evacuate from that place to
used to collect diverse time duration data [18].
safe region.
The time it takes to evacuate is measured to ensure that it
occurs within a reasonable length of time. The duration of
alarms timing is the factor will help to identify the type of
emergency situation in that place to the workers presents in
company. The duration of alarm sign can different for each
events such as pressure release and fire occurs in any place,
sudden temperature increased in any boilers.
At the emergency time, the protection may get fail due to
any technical reason. The number of workers or people died
because of not getting proper training to evacuate that place
at the time of emergency situation.
The mock fire drill should be contacted in all company
with the help of trained people. It must be followed in all
education institution as like school and college, industries.
The factory act 1948 has insisted to all factory should be
contact mock fire drill in some periodical time interval. The
people get important of evacuation timing at the time of T-Test asks whether a difference between two group
emergency state by mock fire drill. averages is unlikely to have occurred because of random
The company has contacted one mock drill by the chance in sample selection. One sample t-test is similar to
suggestion of safety officers and trainer as a routine schedule the independent sample t-test. Expect it is used to compare
of protection timing. Every department of workers has one group average value to a single number. The paired t-
participate that drill in different time duration. test is used in each observation in one group is paired with
Speed room section contained maximum possibility of fire related observation in other group. In our data collection is
accident compared to all departments in company. depends on time taken to reach safe region by workers with
Imaginary fire placed in different side of the machine. The combination of two sample [19, 20].
workers have insisted to evacuate the place. The time
duration of each side of workers to reach the emergency The variance is closely related to standard distribution are
door and normal exit has been noted down by stop watch. measured of how spread out a distribution is. In other words,
The behavior of workers changed at the time of fire they are measurement of variability. The variance is
accidents. It has been clearly observed in mock drill [17]. computed as the average squared deviation of each number
from its mean.

The number of sample has been taken is 12. The hypothesis

rules has applied for small scale test either T-Test and F-
Test. the relationship between the two set of sample should
be different is common condition followed in both of t & f
Test. The main condition of T-Test is two set of sample have

238
 14th International Conference on Humanizing Work and Work Environment HWWE-2016
ISBN Number: 978-93-83006-81-6
same mean. F test is also followed with only two set of Hypothesis method has been choosing to identify
sample have same variance [21]. importance among trial 1 and 3. But the result of T-test gave
Set 1 Set 2 that the both trials are applicable to consider for next level of
Number of Sample N1 = 12 N2 = 12 conformation. High priority number is high in trial 3
sample mean X1 = 1.75 X2= 1.58
compare than trial 2. So the trial 2 has been eliminated from
Sample standard S1 = 1.25 S2= 1.23
deviation documentation.
Table 1: Sample data of T-Test The evacuation time should minimum when compared to
normal exit time. In trial 3 has three different location of
Hypothesis Condition: emergency door as north side right and left corners, south
Null Hypothesis H0 = > X1 = X2 side left corner. The times taken to evacuate the section are
Alternative Hypothesis H1 => X1 ≠ X2 85 seconds and 67 seconds, 140 seconds respectively. The
second illusion has very low time duration for evacuation. It
become safe pathway and suitable location of emergency
door in that speed room section. The location of emergency
T= door has been recommended as south side right corner of the
section.
Sx1x2 = VII. CONCLUSION
Evacuation timing should minimum at the time of
emergency situation. Human life is more important than loss
of material and machine damage. Accounting to model 3 the
= duration of evacuation time is 67 second which is lower
evacuation time in samples. The location of emergency exit
has recommended as diagonal opposite direction of normal
exit. The suggestion has been taken as valuable point in
company. This work is currently in company management
= review and government approval.
=
REFERENCES
= 1.239
[1] Qimiao , jinhui wang, shouxiang, Jan L.M hensen, “An optimization
Sx1x2 = 1.24 method for distance between exits of building considering
uncertainties based on arbitrary polynomial chaos expansion” in
T = Reliability engineering and system safety, volume 154, October 2016.
[2] Zhang, Zhi Sheog Xu, Qian-wei Zhao, Yang-yang Liu “A study of
= 0.17/0.506 theoretical calculation method of subway safety evacuation” Procedia
= 0.336 engineering, volume 71, 2014.
[3] Chang liu, Zhan Li Mao, Zhi Min Fu “Emergency evacuation model
T = 0.34 and algorithm in the building with several exit” Procedia engineering
, volume 135, 2016.
Degree of freedom = N-1 [4] Ruggiero Lovreglio , Enrico Ronchi, Daniel Nilsson “An Evacuation
= 12-1 decision model based on perceived risk, social influence and
= 11 behavioral uncertainty” Simulation and modelling practice theory ,
volume 66, August 2016.
Tabulation value from T –Table [5] E.Palazzi, F.curro, B.Fabiano “A critical approach to a safety
T0.95 = 1.796 equipment and emergency time evacuation based on actual
T0.90= 1.363 information from the Bhopal gas tragedy” Process safety and
The Tabulated value has been taken as one tail test data with environmental protection , volume 97, September 2015.
[6] Chen Haitao, Lou Leilei, Qiu Jiuzi “Accidents causes analysis and
both 90% and 95% variance. evacuation countermeasure on the high rise building fire” Procedia
The calculated value is less than tabulated value, so the Null engineering , volume 43, 2012.
hypothesis will be accepted. [7] Wen- Hui Ju “Study of fire risk and disaster reducing factors of cotton
logistics ware house based on event and fault tree analysis” Procedia
engineering , volume 135, 2016.
VI. RESULT AND DISCUSSION [8] Young Sik Yoon, Dong Han Ham, Wan Chul Yoon “Application of
The number of illusion is more than sufficient level. activity theory to analysis of human related accidents method and case
study” in Reliability engineering and system safety volume 13, 2016.
Unsuitable model should be eliminated from this collection
of trials. Priority factor, hypothesis factor and time factor [9] "Garment industries in Bangladesh and Mexico face an uncertain
future". Textiles Intelligence. 15 October 2003. Retrieved 7 August
were used to remove the incompatible model in group. 2009.
Among 3 groups, trial 2 contains low priority person count is [10] Ethirajan Anbarasan "Dhaka Bangladesh clothes factory fire kills
larger than high priority person count. Trial 2 has take away more than 100". BBC.Archived from the original on 25 November
from list. 2012. Retrieved 25 November 2012

239
 14th International Conference on Humanizing Work and Work Environment HWWE-2016
ISBN Number: 978-93-83006-81-6
[11] Damien Gayle "Bangladesh factory death toll could reach 1,400:
Primark says it will compensate families of killed and injured". Daily
Mail (London) (30 April 2013).
[12] Documented by Solidarity Center, Bangladesh Office “Garment and
Textile Factory Fire Incidents Since Tazreen Fashion Factory Fire”,
November 24, 2012.
[13] Rakesh dubbudu “fire accidents caused an average of 62 deaths per
day in the last 5years” factly.in. april 13, 2016
[14] National Crime Record Bureau ministry of home affairs “Accidental
death and suicide in India 2014 annual report” http://ncrb.gov.in ,
[15] Firoz A “ Design of readymade garments industry for fire safety”. MS
dissertation, BRAC University. (2011)
[16] A. Graziano, A.P. Teixeira, C. Guedes Soares “Classification of
human errors in grounding and collision accidents using the TRACEr
taxonomy” Safety Science 86 (2016) 245–257
[17] Raphael Moura, Michael Beer, Edoardn Patelli, John Lewis, Franz
Knoll “Learning from major accidents to improve system design”
Safety science ,Volume 84, page no (37-45), 2016.
[18] Milad Haghani, majid sarvi “Human exit choice in crowed built
environment: investigating underlying behavioral difference between
normal egress and emergency evacuation” fire safety journal –volume
85, 2016 p.p(1-9)
[19] Junyong Park, Bimal K. Sinha, Arvind K.Shan “Testing interval
hypotheses for scale parameters in gamma distribution” statistics and
probability letter 83, 2013.p.p (2172-2178).
[20] T.Do,M.Haghani ,M.Sarvi, “Group and single pedestrian behavior in
crowd dynamics”, Transp.Res.Rec.:J.Transp.Res.Board(2016)13–19.
[21] Carsten Allefeld, Kai Görgen, John-Dylan Haynes, “Valid population
inference for information-based imaging: From the second-level t-test
to prevalence inference”, Neuro Image Volume 141, 1 November
2016, Pages 378–392.

240

View publication stats