1 Introduction
College dormitories have always been a key unit of safety management for intensive personnel and complex hazards. According to relevant statistics, almost all colleges and universities have had accidents in our country.
In colleges and universities where fires and other accidents have occurred, in addition to destroying the infrastructure, most of them are accompanied by the tragedy of casualties, for example, on February 19, 2003, a fire broke out in the student dormitory of Wuhan University, and 22 dormitories were severely burned. On November 2, 2005, the third floor of the student dormitory building No.6 of Beijing Forestry University suddenly exploded and caught fire, two graduate students were killed in the fire. On January 11, 2007, the graduate dormitory of Northeast Normal University caught fire and more than 500 students were trapped in it. On November 14, 2008, a major fire broke out in the female dormitory of Shanghai Business School. Four girls fell to their deaths on the spot. On May 15, 2017, a fire broke out in the student dormitory of Tianjin University of Traditional Chinese Medicine, and more than 200 students were trapped. In September 2020, a fire broke out in the dormitory in the southern district of Shanghai University Baoshan Campus. In October 2020, a serious fire broke out in the university dormitory in Fanyu District, Guangzhou, and many students were trapped. With the expansion of colleges and universities in our country, the number of permanent students in the building has reached an unprecedented number, which also poses a challenge for emergency evacuation in the event of an emergency.
Related researches believed that there were many factors that affected the evacuation efficiency, which could be divided into three aspects: Environmental factors, management factors and individual psychological behavior factors
[1]. Most of these studies mainly discussed the impact of environmental factors on emergency evacuation
[2], such as the size and number of escape exits
[3], the layout of fire-fighting facilities in the building, building architecture
[4]. the layout of the classroom
[5-7] and the layout of the dormitory
[8]. In addition, factors such as emergency command, fire prevention management
[9], students' fire safety awareness
[10], understanding of evacuation routes
[11], attitudes toward emergency evacuation
[12], evacuation psychology and behavioral responses
[13-15] were also considered to be important factors affecting evacuation efficiency. And had an important impact on the results of the evacuation. However, among the research results of psychological behavior affecting emergency evacuation behavior, the influence of dormitory interpersonal relationship is rarely studied.
College dormitory is one of the gathering places for college students. The time students spend in the dormitory occupies a large part of their daily life, and during the rest period of the dormitory, students' awareness of all aspects is weak. In the event of unexpected events such as fire, earthquake, abnormal disturbances are more likely to prolong the evacuation time. Therefore, the safety management of student dormitories in colleges and universities has attracted more and more attention. In order to fully ensure the safety of students, the evacuation strategies and plans of student dormitories must fully consider the external environmental factors in emergency situations and must also comprehensively consider the psychological and behavioral characteristics of students in emergency situations such as fire.
The age of college students is generally 18
25 years old. They are still in the growth stage in terms of psychology, and there have obvious group characteristics in their daily behavior. In the university dormitory life, the dormitory psychological relationship that occurs, develops and establishes between dormitory members in the process of material and spiritual exchanges has become the most common interpersonal relationship among college students, which has an important impact on the daily psychology and behavior of college students. Compared with other interpersonal relationships of college students, dormitory interpersonal relationships have the characteristics of concentration, interleaving and far-reaching influence, and with the increasing diversity and self-personality of college students, the interpersonal relationship problems of college dormitories are also becoming more and more complex
[16]. Although some studies have mentioned the impact of college students' social relationships on their safety management, most of them are qualitative descriptions, and they have not thoroughly explored the impact of dormitory interpersonal relationships, an important social relationship, on emergency evacuation behavior in emergencies.
2 Research Design
2.1 Behavioral Experiment and Questionnaire Survey
On November 19, 2020, XX University carried out an emergency evacuation drill to enhance students' ability to respond to emergencies, during which an alarm was sounded, and smoke was created to increase the authenticity and tension of the environment. This study took this exercise as an opportunity to carry out behavioral experiments. We recorded the whole process of the experiment. During the evacuation exercise, the evacuation rate was calculated by counting the number of people escaping per unit time. A questionnaire survey was conducted on the students who participated in the evacuation experiment. Three days before the exercise, the "College Student Interpersonal Relations Survey Questionnaire" was distributed to study the characteristics of college students' interpersonal relationships in the dormitory. After the exercise, a questionnaire was issued to investigate the actual follow-up behavior of college students during the evacuation process. 144 questionnaires for girls and 136 questionnaires for boys are used as the original sample data, and the questionnaires were all valid questionnaires. Since students living in the dormitory building have both dormitory interpersonal relationships and the influence of interpersonal relationships such as classmates, this study fully considers the influence of different interpersonal relationships. Before the evacuation drill, the interpersonal relationships of college students were mainly investigated in which interpersonal relationship in dormitory is not separately reflected. The main problems before the evacuation drill are as follows:
"How is your interpersonal relationship (fill in the interpersonal relationship scale)?", "Which classmates do you usually have a better or worse relationship with (write the names of the classmates and rank them)"?
In this emergency evacuation of the dormitory, we found that the interpersonal relationship of college students is mainly reflected in the small group behavior of the dormitory as the unit. During the evacuation, the college students mostly keep the dormitory as the unit of small groups. The main purpose of this study is to explore how this "small group behavior" affects emergency evacuation. Therefore, try to reflect the changes in the relationship between the dormitories of college students after the experiment is completed. The main problems after the evacuation drill are as follows:
"Please recall how many people did you escape with during the evacuation drill just now? Which classmates were they?" (The names of your companions need to be written), "Which dormitories are these students in?" (Check the door number of the dormitory according to their names No). This survey issued questionnaires to all students who participated in the emergency drills, and the questionnaires were filled out following the principle of seeking truth from facts.
Figure 1 A snapshot for the university dormitory emergency evacuation drill |
Full size|PPT slide
2.2 Social Measurement Method and Interpersonal Network
To visualize the structure of interpersonal relationships in college student dormitories, this study uses the Morino psychosocial measurement method (Moreno). It is a method used in sociometric to study the relationship between members of a group. It is mainly used to detect and describe and evaluate people's social status, group structure, and predict group behavior trends by measuring the avoidance tendency of people in the group
[17]. This research is mainly divided into the following two stages:
1) Fill in the social measurement matrix. Number the respondents one by one, summarize the questionnaires, and fill in the matrix. When filling in, the respondent in the questionnaire is used as the starting number to indicate the sender of the relationship, and the selected person is used as the ending number to indicate the recipient of the relationship. In a group, each member is either a chooser or a selected one. The numbers in the matrix represent the relationship between the two. For positive relationships, use the "+" sign, and for negative relationships, use the "-" sign. The value of the number represents the strength of the relationship between the two. According to the survey results, it can be classified into 3 levels. The row and column total numbers represent the total number of edges that have a relationship. Take the matrix table of a 4-person dormitory as an example, the marking method is shown in Table 1.
Table 1 Social measurement matrix table of dormitory interpersonal relationship |
| Member | ❶ | ❷ | ❸ | ❹ | Row Total |
| ❶ | | +2 | -1 | | 2 |
| ❷ | +2 | | | | 1 |
| ❸ | +1 | | | | 1 |
| ❹ | -1 | | | | 1 |
| Column Total | 3 | 1 | 1 | 0 | |
There is a positive second-level attraction relationship between No. 1 and No.2. Find the row of No. 1 in the start number column, find the column of No.2 in the end number column, and then at the intersection of row and column write"+2" above. If there is a relationship of exclusion degree of 1 between No. 1 and No. 3, find the row where No. 1 is in the start number column, and find the column where No. 3 is in the end number column. And then write "" at the intersection of rows and columns. If there is no choice relationship, use blanks.
2) Draw a diagram of the structure of interpersonal relationships. The next step is to draw a diagram of the interpersonal relationship structure of college students according to the social measurement matrix, and use numbers to represent individual members in the group. The solid line between members indicates that the two are attracted to each other, indicates one-way attraction, the dashed line indicates mutual repulsion, and indicates one-way repulsion. The thickness of the line represents the strength of the relationship between the two. Take student No. 1 in Table 1 as an example, draw the relationship between him and the members of the same dormitory, as shown in Figure 2.
Figure 2 The interpersonal relationship structure |
Full size|PPT slide
2.3 Simulation Experiments Based on Social Force Model
In order to investigate in depth the influence of interpersonal relationships in college dormitories on the efficiency of emergency evacuation, this study conducted simulation experiments based on evacuation drills and repeatedly simulated the evacuation behavior of college students under the influence of different types of interpersonal relationships in dormitories by improving the social force model. First, each student is regarded as a self-driven particle whose behavior is the result of the interaction between self-driven forces and the environment
[18], Setting the social force on them to
, the model is as follows:
where is the student's self-driven force, is the inter-student force, and is the interaction between the student and the obstacle. and represent the mass and speed of student respectively, and the acceleration when fleeing is influenced by the individual's self-driven force, inter-individual and inter-obstacle. Let and be the desired escape speed and actual travel speed of the individual, respectively, and be the emergency reaction time. consists of the "psychological force" of the student's instinctive desire to keep a certain distance from other students and the "actual contact force" between students . Where is the distance between the two centers of gravity. is the pointing vector between the students, and is the tangential velocity difference between the two. where represents the tangential direction between the students and the obstacle. All of these main parameters were obtained by this evacuation experiment measurement.
3 Analysis of Dormitory Interpersonal Relationship and Evacuation Following Behavior
3.1 Normal Interpersonal Network Structure
First, this paper analyzes the interpersonal relationship structure of the sample college students in their daily lives, divides the sample students into groups according to gender. Among them, girls are No.1144, and boys are No.1136. First, we measure the students' interpersonal relationship selection tendency based on the results of the interpersonal relationship questionnaire survey of college students before the exercise experiment. By filling in the social measurement matrix and using Gephi to draw a diagram of the interpersonal relationship structure, the solid line represents attraction, and the dotted line represents repulsion. The normalized interpersonal network structure of the 280 students participating in the experiment in 70 dormitories is shown in Figure 3. The black nodes represent the junior-grade students (first and second grades), and the white represents the senior-grade students (third and fourth grades).
Figure 3 Student's interpersonal relationship network in normal condition |
Full size|PPT slide
Through the network structure, it can be found that there are obvious characteristics of interpersonal relationships among college students in small groups based on dormitories, and there is also a certain cross-dorm cross relationship. By comparing the results of the questionnaire survey, this study found that the interpersonal relationship of college students in the dormitory can be roughly divided into three types: Unity, loose and opposing. Unity type refers to the members of the dormitory are as harmonious and intimate as the whole. Loose type refers to the contradiction among individual members of the dormitory, and there are small groups within. Oppositional type refers to the mutual exclusion of dormitory members, often acting alone
[16]. To quantify the types of dormitories, let Ts represent the type of dormitory,
is the number of people in the dormitory, and
is the number of interpersonal network relationships in the dormitory. According to the clustering algorithm:
Through statistics, it is found that the proportion of interpersonal relationship types of sample college students in the dormitory under normal conditions is shown in Table 2.
Table 2 The proportion of three types of dormitory interpersonal relationships in normal |
| | Object | Relationship types |
| | | Oppositional | Loose | Unity |
| Boy | Junior grades | 11.76% | 29.42% | 58.82% |
| Senior grades | 5.82% | 41.18% | 52.94% |
| Girl | Junior grades | 11.11% | 27.77% | 61.11% |
| Senior grades | 22.22% | 22.22% | 55.56% |
| Total | | 12.86% | 30.00% | 57.14% |
It can be seen that most of the dormitory relationship types are unity, and there is no significant difference in this dormitory relationship among students of different grades and different genders. In oppositional and loose dormitories, there are certain differences between different grades and different genders. Compared to junior grade, senior-grade girls are more likely to form opposite-type dormitory relationships, and senior-grade boys have more loose dormitory relationships. Also, more in line with the characteristics of interpersonal relationships of contemporary college students
[16]. Therefore, in the process of studying the influence of interpersonal relationships, it is necessary to consider the differences in gender itself, and at the same time the influence of grade growth on the psychological changes of students.
This study found that the interpersonal relationship of the sample college students under normal conditions is dormitory roommates as the main body. Although there are interpersonal relationships between dormitories, there is no obvious opinion leader. The cross-relationship between the dormitories due to the relationship between classes and friends may further reduce or expand the size and the overall number of small groups in individual dormitories. This kind of social relationship behavior can also easily cause students in different dormitories to follow each other, which may increase the uncertainty in the evacuation process
[18].
3.2 The Network Structure of "Leader-Follower Relationship" Under the Emergency State of College Students
Different from normal interpersonal relationships, when emergencies break out, people enter a state of tension, and in the process of evacuating and fleeing, a different interpersonal following relationship may occur
[19], that is, the "herd effect'' is formed. In order to further explore the actual escape behavior of college students in the state of emergency evacuation, we fill in the social measurement matrix table and draw the interpersonal relationship structure with "please recall which classmates you followed to escape during this evacuation exercise?" as the selection criterion, the result is shown in
Figure 4.
Figure 4 Student's follow-up relationship network during emergency evacuation |
Full size|PPT slide
It is obviously different from the interpersonal relationship under normal conditions, from the follow-up relationship structure in
Figure 4, there are obvious "opinion leaders" among the students during the actual emergency evacuation process, such as the 10th, 31st and 51st girls in the lower grades. No. 84 and No. 117 in senior grades are both cadres of the student union; No. 23 and No. 46 for boys in lower grades are class cadres, No. 60 is an cadre of the student union, and No. 60 is a cadre of the student union. No. 86 and No. 106 are student union officials. Therefore, there will be obvious "leadership-following behavior" among college students in the state of emergency
[20, 21], in which the central role of class cadres and student union cadres is very significant, and this effect is more pronounced in junior grades than in higher grades. In this state, a certain "substitute relationship" will be formed for the change of the dormitory relationship.
In addition, through an intuitive comparison with the normal interpersonal relationship structure, the degree of interpersonal connection between male and female dormitories has decreased to varying degrees in an emergency state. The value is still used as the basis for dividing the dormitory relationship type, lowercase letters and uppercase letters are used to number the dormitories of junior and senior grades respectively. Count the proportions of the types of interpersonal relationship between the female and male dormitories in the evacuation state and compare the changes in the degree of closeness of the dormitory relationship in the two states, as shown in Figure 5 and Figure 6.
Figure 5 Closeness of interpersonal relationship in female dormitory |
Full size|PPT slide
Figure 6 Closeness of interpersonal relationship in male dormitory |
Full size|PPT slide
It can be found that the closeness of interpersonal relationships in the dormitories of the sample students will be significantly reduced in an emergency state, and the proportion of different types of dormitory relationships has also changed significantly. In the state of emergencies, a considerable number of dormitory relationships have changed to opposing and loose relationships. It also reflects the tendency of college students to escape alone and follow "opinion leaders" in an emergency escape state. In order to quantitatively analyze the influence of the normalized dormitory relationship on emergency evacuation behavior, the parameter is introduced, make
Among them, is the number of overlap between the total number of follower relationship edges in emergency and the total number of interpersonal relationship edges in normal, is the total number of interpersonal relationship edges in normal. Use dormitory as a unit to explore the values of different grades, different genders, and different types of interpersonal relationships, and use the test for the differences in different groups. The influence of the normalized dormitory interpersonal relationship of college students on their emergency follow-up behavior can be calculated, and the results are shown in Table 3.
Table 3 Influence degree of student dormitory interpersonal relationship on the following behavior |
| | Influence | Type of dormitory | |
| | | | Oppositional | Loose | Unity |
| Ir | Boy | Junior grades | 0.32 | 0.63 | 0.79 | 14.01*** |
| Senior grades | 0.29 | 0.49 | 0.56 | 13.18*** |
| 7.26* | 10.23** | 9.59** | - |
| Ir | Girl | Junior grades | 0.34 | 0.66 | 0.88 | 13.05*** |
| Senior grades | 0.33 | 0.50 | 0.63 | 12.24*** |
| 8.05* | 10.18** | 7.78** | - |
| Notes: * is the significance coefficient , ** is the significance coefficient , *** is the significance coefficient . |
Through comparison, it can be found that the normalized interpersonal relationship in the dormitory has a higher impact on the evacuation behavior of girls than boys, and the impact of junior grade students is higher than that of senior grade students. In addition, there are also certain differences between different types of dormitories. The unity dormitory is most affected by the interpersonal relationship in the dormitory, and the proportion of closely following relationships formed in the state of emergency evacuation is greater. Opposite dormitories are the least affected and are most likely to form independent escape or "leader-following" behavior under emergency conditions; loose dormitories are somewhere in between. Therefore, the influence of interpersonal relationships in college dormitories on their following behavior under emergency evacuation cannot be ignored. Generally speaking, in the process of emergency evacuation, although students of different grades and genders have different behavioral tendencies, they will still be affected by the normalized interpersonal relationship in the dormitory to varying degrees. The actual evacuation of college students is still mostly carried out in dormitory units. To explore the impact of this relationship on the actual evacuation efficiency, this study further attempts to conduct a comparative study through behavioral experiments.
4 The Influence of Interpersonal Relationship in Dormitory on Evacuation Efficiency
4.1 Behavioral Experiment Under the Influence of Normal Dormitory Relationship
Through the above research we find that the interpersonal relationships of college students will change significantly under normal and emergency conditions, and this is characterized by changes in dormitory relations. Although students of different grades and genders have different behavioral tendencies, but affected by the normalized dormitory relationship, the actual evacuation of college students can still largely carry out in small groups in dormitories. In order to explore the impact of different interpersonal relationships, especially the changes in dormitory interpersonal relationships, on emergency evacuation, this study uses the "pedestrian and social force module" in the AnyLogic simulation software to reproduce different scenarios of dormitory emergency evacuation and simulates changes of emergency evacuation in different scenarios.
AnyLogic simulation mainly realizes the effect of interpersonal relationship on crowd evacuation behavior by setting the group size and number of groups of agents. For the parameters of the model, we set it based on the results of the survey and combined with the social force model. Therefore, this study is based on the interpersonal network of college students obtained by the survey (Figure 3 and Figure 4), and the model is set with the same proportion of students as the survey results, and the follow-up evacuation are carried out according to the interpersonal network relationship. For example, if there is interpersonal relationship or following behavior between the two, the evacuation experiment is performed by setting the two to keep a certain distance in the simulation.
For the influence of interpersonal relationships among college students, through repeated observation and measurement of the evacuation process, it can be found that the distance between the individual centers of gravity of two students with one-way attraction is m, and the distance between two students with two-way attraction is m, and the distance between mutually exclusive groups has no obvious regularity.
In order to reproduce the realistic results more realistically, the initial personnel layout and dormitory distribution of the simulation were kept consistent with the evacuation exercise. Firstly, the simulation environment was established based on the actual parameters of the dormitory of XX University, which has one entrance and four staircases, and 32 dormitory rooms on each floor. When an emergency occurs, the exit of the dormitory building will be completely opened, and all people will follow the nearest route to escape quickly to the safe area, the simulation environment is shown in Figure 7.
Figure 7 Emergency evacuation model under the influence of heterogeneous individuals |
Full size|PPT slide
Through the actual measurement of this emergency drill, the average escape speed of female students is about 1.45 m/s and male students is about 1.65 m/s, and the speed of moving downward on the stairs without the influence of obstruction is 0.8 m/s and 1.1 m/s respectively. Combining Helbing's findings with the empirical results of this study to iteratively calibrate the social force model
[19, 22], the social force parameters in the simulation were set to
N,
m,
,
,
s, The evacuation process makes the expected value of individual college students' escape speed
. In the process of modeling, we repeatedly compared with the evacuation video and data to ensure that the simulation is close to the real movement state of college students.
4.2 Analysis of Results
We record the emergency evacuation efficiency of students of different genders and different grades in the three states respectively, and obtain the escape rate at the exit, as shown in Figure 8.
Figure 8 Emergency evacuation speed under the influence of dormitory interpersonal relationship |
Full size|PPT slide
From the experimental results, it can be seen that whether girls or boys, the escape rate is: independent evacuation higher grade emergency higher grade normal junior grade emergency junior grade normal. Overall, the junior grades were evacuated faster, and the results of the three evacuations were the same. Furthermore, this research has two important findings: 1) When students escape independently, the overall evacuation efficiency is the lowest. The close interpersonal relationship in the dormitory can significantly improve the evacuation efficiency. In the behavior experiment, the evacuation rate under the influence of interpersonal relationship is higher than that of independent evacuation without the influence of interpersonal relationship. 2) The interpersonal relationship structure formed in the dormitory under normal conditions has better evacuation efficiency than the "half opinion leader-half dormitory small group'' relationship structure formed in an emergency state, and this is particularly obvious among junior grade students. In contrast, senior students are more likely to evacuate relatively alone, and the evacuation speed is slower than the junior students.
The above results are based on the premise that there is no obvious congestion. During the experiment, we measured the spatial density of the crowd to stabilize at 0.72 to 1.08 people. In the case of a student falling, during the evacuation process, the left passage and stairway of the dormitory caused crowds to overlap each other and congested. At this time, the influence of the interpersonal relationship in the dormitory on the efficiency of emergency evacuation is no longer significant (the results of this experiment are invalidated). Therefore, emergency evacuation is the result of the interaction between people and the environment. The adaptability of dormitory escape routes and the number of student groups is the key, and it is also the prerequisite for the influence of dormitory interpersonal relationship on evacuation efficiency.
5 Conclusion
1) Conduct research on emergency evacuation behaviors of college students through behavioral experiments and investigations, we obtain more real experimental data, and find that there are three dormitory interpersonal relationships among college students' dormitory relationships: Oppositional, loose, and unity. And there are some differences among the three dormitory relationships among different grades and different genders. This relationship will have an important impact on emergency evacuation in emergencies.
2) The normal interpersonal relationship structure of college students has obvious small group behaviors based on dormitories. Junior grades have a higher degree of closeness in dormitory relationships than senior grades. In an emergency, the tightness of this dormitory decreases significantly. The normal dormitory relationship has a significant impact on the follow-up behavior during emergency evacuation. This effect is stronger for girls than for boys, and junior grades are stronger than senior grades.
3) The close united dormitory relationship can significantly improve the evacuation efficiency, which shows that the emergency evacuation state of small groups following and mutual assistance in the dormitory unit has a higher evacuation efficiency and can effectively avoid congestion and trampling caused by falling behind, chaos, etc. behavior. In an actual emergency, this relationship will be significantly weakened and partly replaced by a follow-up relationship based on "opinion leaders", which actually increases the disorder and complexity of evacuation and is not conducive to rapid evacuation.
4) The behaviors of students of different grades and genders are obviously different during the evacuation process. Among them, the junior grades have a closer and orderly relationship with the dormitory compared to the senior grades, and the evacuation process is faster. During the evacuation process, female students are relatively more affected by the interpersonal relationship in the dormitory. Therefore, in safety education, appropriate intervention and guidance should be carried out for students of different grades and different genders.
{{custom_sec.title}}
{{custom_sec.title}}
{{custom_sec.content}}
PDF(1441 KB)
Figure 1 A snapshot for the university dormitory emergency evacuation drill
Table 1 Social measurement matrix table of dormitory interpersonal relationship
