Formation mechanism of quick emergency response capability for urban rail transit: Inter-organizational collaboration perspective

Abstract

With the rapid development of urban rail transit in China recently, improving its quick emergency response capability is becoming an important issue. Based on the perspective of inter-organizational collaboration, this article examines the formation mechanism of quick emergency response capability of urban rail transit and proposes the concept model hypothesis, in order to highlight the inter-organizational emergency collaboration relationships and the quick emergency response capability. According to site surveys and analysis of the elements of inter-organizational collaboration in emergency rescue and the meaning of quick emergency response capability, the scale of emergency collaboration and emergency response capability is designed, and the hypothetical concept model is tested by structural equation model. The results indicate that the emergency collaboration can be realized mainly through emergency organizations, resources, plans, and information. These elements interact with each other; the quick emergency response capability includes fast reaction and emergency disposal capability, emergency decision and execution capability, and coordination and joint action capability. These capabilities restrict each other. Moreover, emergency collaboration has significant but different influence on different dimensions of quick emergency response capability. Therefore, allocating and controlling emergency elements are pivotal to realizing inter-organizational emergency collaboration and generating the quick emergency response capability of urban rail transit.

Keywords

Urban rail transit transport engineering emergency inter-organization collaboration quick emergency response capability structural equation modeling

Introduction

In recent years, frequent underground emergencies worldwide have led to many casualties and property loss, as well as unfavorable political influence on public security and stability, leading to social anxiety. Disaster or emergency management has attracted contributions from researchers all across the globe.¹ Moreover, it must be acknowledged that emergency management plays a vital role in minimizing loss if immediate after an accident. In this article, we discuss emergency response, regarded as the process of gathering resources and acting immediately after the incident happens.² The process is from the occurrence of a threat or an incident to responses and actions, and the goal of emergency response is to help rescue injured personnel, protect vulnerable people, and control and lessen the impact of an accident as much as possible.³

We can infer that most disaster relief organizations are trying to move relief goods and rescue people more quickly and effectively,⁴ thus emergency response must be agile, resilient, and active. Similarly, we come across words such as agility, flexibility, and so on in the literature of humanitarian relief and supply chain.^5–7 As the frequency of natural disasters and underground emergencies have vastly increased in recent years, the pressing need for emergency response remains an issue despite increasing contributions in the field.

Emergency response is essential to achieving effective emergency management. One of the most important components of emergency management is to evaluate the emergency response capacity of an emergency department.⁸ Emergency response capability (ERC) is a comprehensive capability to deal with natural disasters, sudden public health incidents, sudden public safety incidents, military conflicts, and so on.⁸ If some emergencies occur, adequate resources would be allocated to relief, a huge number of measures would be taken in time through the emergency response capacity. Besides, proper information collection, sharing, cooperative communication, and processing mechanisms would lead to effective coordination and efficient operational outcomes.⁹ Information sharing and coordination during inter-agency response also improve effectiveness of response.¹⁰ Emergency response requires collaboration between numerous people and groups.¹¹ ERC would strengthen as a result of coordination or collaboration across responding organizations.¹² However, detailed investigations into the effect of emergency collaboration (EC) on ERC are lacking.

Consequently, we derived our research objectives to develop ERC based on inter-organization collaboration perspective. To further achieve our research objectives, this study applies the resource-based view (RBV),¹³ information-processing view IPV,¹⁴ and coordination theory to help our understanding of how organizations can create the quick ERC. The RBV argues that “organizations may achieve competitive advantage that bundling of resources to create capabilities.”⁵ IPV advocates that “the greater the task uncertainty, the greater the amount of information that must be processed among decision makers during task execution in order to achieve a given level of performance.”¹⁵

To answer our research objectives, this study aims to investigate the associations between ERC and EC of urban rail transit in China through a positivism approach, such as structural equation model (SEM). This article is therefore organized as follows. Section “Theoretical foundation and hypothesis development” focuses on concept model building and hypothesis formulation, including a review of related research, defining emergency response and EC. Section “Research methods” deals with questionnaire design and data collection. Section “Test of hypotheses and results” deals with sample assessment, hypothesis testing and shows the analysis results. Section “Discussion and limitations” discusses the empirical results, limitations, and further research directions.

Theoretical foundation and hypothesis development

Theoretical framing

The complex and dynamic nature of emergency response can be influenced by the large number of different factors in the changing environment. The RBV asserts that organizational capacities are defined as a higher-order construct which relies on the bundling of resources.¹⁶ The bundling of resources is necessary to capabilities formation,¹⁷ if resources are combined and utilized together, they could create capabilities. Resources can be categorized as physical capital, human capital,¹³ and extended as financial capital and technological capital.¹⁸ When resources are processed or utilized in bundles, the integration of resources could drive performance and develop capability.¹⁹

The IPV suggests that in the wake of disaster or emergency, uncertainty is high, quick, and accurate information is necessary.¹⁵ When the environment is rapidly changing, units are interdependent, prompt information delivery is critical.²⁰ Efficient exchange of information could reduce uncertainty in the changing environment.²¹ In IPV, organizations should adopt coordination to enhance the information processing.¹⁴ Besides, organizations link through the channels of communication among units to facilitate effective coordination.²²

In the humanitarian relief and emergency management, information flows and adequate resources provide such a platform for coordination/collaboration among peer emergency organizations.²³ Moreover, there are many other factors affecting coordination in emergency, our study focuses on the formation of EC leads to a better emergency response and creates ERC. Thus, we posit that emergency organization, resource, plan, and information directly affect EC to build the emergency response capacity. Emergency organizations, emergency resources, emergency plans, and emergency information interact with each other, realizing EC and generating quick ERC, as shown in Figure 1. On the basis of evidence of extant literature and theory, this framework model describes the elements of inter-organizational EC and the relationships between inter-organizational EC and quick ERC. In the following subsections, we explain the components of our proposed framework and introduce our hypotheses.

Figure 1.

The proposed framework for emergency response.

EC

Collaboration has been and remains a significant obstacle to effective emergency management.^24,25 In recent years, some scholars have done contributions about the factors influencing or relating to EC, as shown in Table 1. McEntire²⁶ in a case study examined how organizations collaborate in an attempt to perform multiple response and recovery functions, suggested a number of reasons to show why emergency operations proceed smoothly, including political support, preparedness measures, networking, technology, cooperative relationships, and so on. In addition, access to core information enhances the efficiency of response actions and increases coordination throughout the network of responding organizations.²⁸ Likewise, Robinson et al.³² explored the factors which influence the development of stable emergency response collaborative partnerships and emphasized the importance of communication and interaction in the development of disaster response networks. Kapucu²⁹ found that pre-season planning, open communication between emergency managers and elected officials, and the use of technology all had a significant impact on community responses.

Table 1.

The extant literature of factors related to emergency collaboration.

Literature	Elements
McEntire²⁶	Political support, preparedness measures, networking, technology, and cooperative relationships
Waugh²⁷	Decision-making styles, organizational cultures
Comfort et al.²⁸	The type and amount of information available, the degree of planning, preparedness prior to the event, the specific time, location, and magnitude of the incident, and the existing organizational resources or constraints
Kapucu²⁹	Pre-season planning, open communication, elected officials and the use of technology
McGuire and Silvia³⁰	Problem severity, organizational and managerial capacity, and internal structure
Wang³¹	Task importance, resources shared by response goals, and key relations for each task

Extensive research has been done on the influential factors of inter-organizational EC. By analyzing the case of Fort Worth tornado on 28 March 2000 in the United States, researchers suggested certain factors inhibited emergency coordination, that is, information challenges, a lack of communication between the field and emergency operations center, equipment failures, language barriers, and a command and control mentality.²⁶ Waugh²⁷ argued that emergency organizations typically had their own decision-making styles and differences in organizational cultures had tremendous impact on how emergency organizations interact with other organizations outside and with individuals. The efficiency of EC is related to the type and amount of information available; the degree of planning and preparedness prior to the event; the specific time, location, and magnitude of the incident; and the existing organizational resources or constraints.²⁸ In analysis of four hurricanes cases between 13 August and 25 September 2004, in Florida, USA, researchers found that emergency communication procedures available and utilized, a plan to alert all agencies of a threat, and the use of information technology to improve communication and coordination among agencies were effective coordination strategies.²⁹ Additionally, McGuire and Silvia³⁰ studied the efficiency of EC from three aspects: problem severity, organizational and managerial capacity, and internal structure. Likewise, Wang et al.³¹ found out information in three aspects, named task importance, resources shared by response goals, and key relations for each task, contributed to effective collaboration during the process of emergency response.

Based on the research achievements and by merging and classifying the influential factors of inter-organizational EC, this article presents the inter-organizational EC elements of urban trail transit. These elements include emergency organizations, emergency resources, emergency plans, and emergency information.

Emergency organizations

Emergency organizations are considered the core of emergency management systems. These constitute the main body for building daily emergency management systems and supervising rules and regulations, as well as commanding executive agencies. Internationally, emergency organizations are considered the omnibearing, multi-agent, multi-level, and comprehensive organizational networks for disaster response, which include government, police, fire services, medical services, media departments, and private organizations, rather than a single organization or several organizations that participate in building emergency management systems in emergency response without coordination. Emergency organizations are the most active elements of the quick ERC of urban rail transit, closely related to emergency resources, emergency plans, and emergency information.

Emergency resources

Emergency resources refer to the various essential resources for effectively undertaking emergency activities and guaranteeing an emergency rescue system to run normally in an emergency response process for urban rail transit. Also, emergency resources are the foundations upon which to develop the quick ERC of urban rail transit, including human resources, goods and equipment, funds, technologies, emergency facilities, historical data, laws and regulations, and so on. Rational reserves and deployment of emergency resources are fundamental for quick ERC.

Emergency plans

Emergency plans aim to prevent and cope with potential incidents or disasters. This means that the several schemes prepared in advance or arrangement for guaranteeing emergency rescue in a quick, orderly, and effective manner, to minimize accidental losses, are the most crucial elements of the ERC of urban rail transit. Emergency plans which are formulated based on different kinds of emergencies could not only ensure emergency action as planned and realized quickly and with high efficiency but also provide guidance for daily training and drills for emergency rescuers as well as maintain emergency resources in a good state. The activities include formulating, exercising, evaluating, and revising emergency plans.

Emergency information

Emergency information refers to the processed data such as figures, images, texts, and so on in the emergency response process for urban rail transit. This is the important foundation of emergency decisions and emergency rescue. Moreover, the quick ERC of urban rail transit depends largely on the timeliness of mastering emergency information. Emergency information is closely related to the emergency management information system (EMIS). EMIS is an emergency information exchange platform for urban rail transit, which employs multiple means and methods, from multiple perspectives, to gather and transfer data and messages. Effective exchange of emergency information could improve the coordination of emergency organizations to realize the high efficiency of organizational management, and to make emergency organizations work in cooperation with the external environment to raise the strain capacity for handling emergencies. Consequently, we propose the following hypothesis:

Hypothesis 1. EC is composed of four elements, that is, emergency organizations, emergency resources, emergency plans, and emergency information.

ERC

Emergency response has attracted much research attention in recent decades. Researchers have generated a vast literature covering a broad range of objectives, perspectives, and levels of analysis. Emergency response research began in the wake of the Three Mile Island Nuclear incident in Pennsylvania, USA, in 1979.³³ This incident brought about sweeping regulatory changes in emergency response practice as well as elevated research activities in this area.³⁴ Since then, researchers have studied emergency response related to other natural disasters such as hurricanes, earthquakes, and wildfires.^35,36 Emergency response demands fast and effective action, often in life-threatening situation.¹¹ Additionally, it includes alarm systems, accident notification, accident reporting, command decisions, resource allocation, coordination, rescue emergency shelter, expert support and evacuation, and so on. Procedure planning is an important issue in the emergency process; an emergency plan serves both as a legal document and as a manual that describes those procedures. An emergency plan contains information about the kind and location of available safety equipment (such as emergency exits and fire barriers) and the detailed procedures to follow in a number of eventualities.³⁷ Researchers argue that an effective emergency rescue plan (ERP) enables cooperation and coordination among the human agents involved.³⁸ Bruyelle et al.³⁹ emphasize the importance of the origin of the message in the process of emergency decision-making and suggest communication is important for informing the authorities of the situation and expediting the organization of the rescue during the process of emergency response. Furthermore, resources are essential for emergency rescue. The emergency response team (ERT) is also considered the most effective approach for dealing with emergencies in industries, and for minimizing the risk of casualties and losses. Several groups and team members with different levels of experience, roles, and responsibilities work together in the ERT.⁴⁰

The United Nations Development Programme (UNDP) defines capacity as the ability of individuals and organizations or organizational units to perform functions effectively, efficiently, and sustainably.⁴¹ The capacity assessment model proposed by the UNDP shows obvious advantages compared with other capacity models, which means a structured and analytical process whereby the various dimensions of capacity are measured and evaluated within the broader environmental or systems context, as well as specific entities and individuals within the system. According to Hu et al.,⁴² the framework for emergency response capacity consists of three levels: systems, organizational, and individual. The systems level contains policy, regulations, process, and so on, and the organization level consists of strategy, culture, resources (human, financial, information), and so on. Some researchers^43–46 have also been involved in exploring the essential problem of ERC evaluation. There are two modes of ERC evaluation: first, dealing with the emergency incidents in a timely fashion; second, reducing the total costs of emergency rescue and restoration.

Quick ERC derives from the fundamental elements necessary for emergency rescue and is based on integrating, allocating, and applying these elements. The function of quick ERC relies on reacting to emergencies as quickly as possible, disposing emergencies to prevent conditions from worsening, making scientific decisions immediately, and executing effectively. Moreover, it involves coordinating individuals and organizations participating in emergency rescue to work toward the same goal efficiently. Although the emergency proposal of urban rail transit emergencies captures special engineering background, quick and effective decisions, execution and control, and omnibearing coordination are essential for generating quick ERC. Hence, we suggest the following hypothesis:

Hypothesis 2. Quick ERC is composed of three dimensions, that is, fast reaction and emergency disposal capability, emergency decision and execution capability, and coordination and joint action capability.

Impacts of inter-organizational EC on quick ERC

The formation of quick ERC of urban rail transit depends on the implementation of EC from inter-organizations, resources and information, and inter-organizational EC. This could support and promote quick ERC. These approaches, that is, breaking organizational boundary, preparing and quickly implementing emergency plans, effectively allocating resources, and gathering and transferring information, could drive the formation of quick ERC and raise the efficiency and level of emergency management. Thus, we posit the following hypothesis:

Hypothesis 3a. Emergency response collaboration has positive effect on quick ERC (Model I).

Hypothesis 3b. EC has positive effect on fast reaction and emergency disposal capability, emergency decision and execution capability, and coordination and joint action capability (Model II).

Research methods

Measures

If possible, established scales from the extant literature can be utilized in this study.⁴⁷ However, we could not identify suitable measures of EC and ERC of urban rail transit. Instrument development procedures for these constructs followed Churchill’s⁴⁸ scale development methodology including a comprehensive literature review, followed by pre-testing with academics in the field of risk or disaster management. To guarantee the science and effectiveness of the survey, the initial scale was formulated combined with previous research on and the characteristics of urban rail transit, and modified through the expert interviews and pre-investigation, which means we made minor modifications to the wording of items based on the feedback from pre-tests in order to improve scale performance. To measure the four dimensions of EC and three dimensions of quick emergency response capacity, all scales were designed using the 5-level Likert-type scale, with rating from 1 as strongly disagree to 5 as strongly agree. Thus, the respondents could fill out the survey according to their experience and theoretical knowledge. The items were screened to form the formal questionnaire, using a 17-item EC scale and an 11-item ERC scale.

Procedure

As mentioned, prior to data collection, the content validity of the scale was established by grounding it according to existing literature and expert view and conducting pre-tests. Then, we will conduct three-stage instrument development process, namely, reliable analysis, exploratory factor analysis (EFA), and confirmatory factor analysis (CFA) for assessing the construct reliability and validity of the instrument⁴⁹ and conducting the empirical research.

Based on the existing literature, the initial scale was developed and pre-tested combined with expert interviews to determine the formal questionnaire of EC and fast ERC. Then, the questionnaires were sent out and collected through e-mail and online survey. After the data collection, Cronbach’s alpha value was generated for each construct. If Cronbach’s alpha value was greater than 0.7, the construct was accepted.⁵⁰ Moreover, the constructs with an acceptable Cronbach’s alpha value of at least 0.6 were further evaluated for the possible improvement. The item inter-correlation matrix was adopted in determining the items whether the best candidates for deletion. If the items that negatively correlated to other items, they were first discarded. The items with correlation value below 0.1 also were discarded. After the elimination of some items, the reliability analysis was performed again. If the construct still failed to achieve the minimum alpha value of 0.6, it would be deleted. Since all constructs get the target value, the reliability analysis completed and moved on the next analysis.

Second, EFA was conducted through principal component analysis (PCA). The common method of varimax rotation with Kaiser normalization was used to clarify the factors.⁵¹ The exact number of factors would be extracted through EFA using SPSS software. The items would load on factors, determined through comparing their loading on the factors. The items, which load on they intend to measure, could be kept from consideration.

Finally, the last step involved CFA in evaluating construct validity, and path analysis in assessing the relationship among factors. The CFA and hypothesis testing could be conducted through SEM using AMOS software to examine the relationship between EC and fast ERC. Some goodness-of-fit indices (GFIs) were used to evaluate the tenability of models, such as GFI, comparative fit index (CFI), root mean square error of approximation (RMSEA), and so on.⁵²

Data collection

The unit of analysis employed in this study was at the level of the emergency response institution of urban rail transit. The target sample was composed of the professionals who have experience of engaging in urban rail transit emergency response for more than 3 years. We deem these professionals to be the most knowledgeable about emergency response and our related subjects of interest: emergency resources, plan, information, organization, and the capability of quick emergency response.

The hypotheses were tested with data collected from e-mail and online survey. The Internet-based survey in China was utilized for data collection. As mentioned above, a 5-point Likert scale with end points of “strongly disagree” and “strongly agree” was used to measure items. Of the 240 questionnaires, 204 questionnaires were returned completed, with 159 valid responses and a return rate of 77.9%. Preliminary data reveal that 85.6% have a Bachelor degree or above, 13.8% are senior managers, 46.8% are middle managers, 39.6% are operating personnel, and the average working life is 5.5 years. Therefore, although the total samples are reduced, the valid samples are representative and reliable enough to reflect the whole characteristics. Respondent profile is presented in Table 2.

Table 2.

Respondent profile.

Respondent	Count	Percent
Senior managers	22	13.8
Emergency Command Center	4
Rail Transit Group Co., Ltd	11
Health Sector	3
Government	4
Middle managers	74	46.6
Rail Transit Command Center	15
Rail Transit Construction Administration Corporation	19
Railway Administration	23
Social organization	6
Government sector	8
Public Security Bureau	3
Operating personnel	63	39.6
Construction management personnel	28
Medical workers	12
Firemen	11
Government agents	8
Resource blending operator	4

Non-response bias

Non-response bias is the difference between the answers of respondents and non-respondents.⁵³ In this study, non-response bias was assessed comparing the responses of e-mail and online samples of returned surveys to provide support of non-response bias.^53,54 The final sample was split into two, depending on the kinds that they are inquired. The group of online survey consisted of 96 responses, while the group of mail survey consisted of 63 responses. The t tests performed on the responses of these two groups yielded no statistically significant differences (at 99% confidence interval). The result suggests that non-response does not appear to be a problem.

Test of hypotheses and results

Reliability test

Reliability means the consistency degree of the measurement, or the possibility of the same result when measuring the same object in the same case, reflecting the consistency or stability of a measuring tool.^55–57 Generally, Cronbach’s alpha can be seen as the reliability coefficient, which over 0.80 means great significance.⁵⁸ The reliability analysis result is shown in Table 3. Cronbach’s alpha of the questionnaire is 0.865, and Cronbach’s alpha values of EC and quick ERC are 0.839 and 0.823, respectively. Therefore, all the values are over 0.80 and illustrate the high internal consistency and reliability of the questionnaire.

Table 3.

Reliability statistics.

Constructs	Items	Cronbach’s alpha	Total Cronbach’s alpha
Emergency collaboration	17	0.839	0.865
Quick emergency response capability	11	0.823	0.865

Validity test

Factors analysis of the EC questionnaire

EFA examines whether a common factor of the biggest variance ratio can be extracted through all the items which, theoretically, belong to the same dimension, in order to conceptualize the connotation of this dimension. Before the factor analysis, it is necessary to test whether the sample is appropriate for factor analysis. The Kaiser–Meyer–Olkin (KMO) measure and Bartlett’s sphericity test can be used to determine the suitability of the sample. Generally, if KMO measure is over 0.7, it is appropriate for factor analysis. Moreover, Bartlett’s test is required to reach the significance level. The method of PCA and varimax rotation algorithm can be used to capture the common factor. The number of the extracting factor is determined through whether the eigenvalues are over 1; the standard of deleting the item is whether the factor loading is less than 0.4; and the cumulative variance of extracted factor can explain over 50%.

The KMO and Bartlett’s test (Table 4) show that the KMO measure is 0.784, and the significance level of Bartlett’s test is less than 0.001. Therefore, the sample of EC is suitable for EFA. Four common factors are extracted; the cumulative variance can explain 58.91%, the table of factor loading is shown in Table 5, and factor loading less than 0.4 has been deleted, and the bigger one is reserved.

Table 4.

Kaiser–Meyer–Olkin and Bartlett's test of Emergency Collaboration.

Kaiser–Meyer–Olkin measure of sampling adequacy	0.784
Bartlett’s test of sphericity
Approx. chi-square	974.761
df	136
Sig.	0.000

Table 5.

Rotated component matrix of emergency collaboration.

Item		1	2	3	4
VAR01	In the process of emergency rescue, my organization can smoothly communicate with other organizations			0.605
VAR02	My organization always quickly matches up to the action of other organizations			0.781
VAR03	Other organizations can quickly act together with my organization			0.869
VAR04	The staff in all organizations are stable in the process of emergency response	0.746
VAR05	There are adequate materials and equipment to cope with an emergency in my organization	0.731
VAR06	My organization can in a timely manner allocate the resources	0.688
VAR07	There are adequate professionals and a high technical level of emergency response in my organization	0.775
VAR08	There is an improved emergency reserve management system in my organization	0.692
VAR09	An emergency joint warning system has been established		0.658
VAR10	There is a scientific, complete, and feasible emergency plan in each organization engaging in emergency response		0.649
VAR11	Organizations always carry out emergency plan training and drills jointly		0.629
VAR12	There is strong connection and complementation of the emergency plan from each organization		0.519
VAR13	Every organization has signed the complete emergency agreement		0.433
VAR14	Information is easy to access in the emergency response process				0.474
VAR15	Information is provided to each organization in a timely and complete manner in the emergency response process				0.722
VAR16	The development trend of emergency gives the feature of uncertainty				0.716
VAR17	Many informatization methods are used to assist in the emergency response process information transfer				0.689

The first-order and second-order CFA are conducted via the real surveyed specimen (n = 159). As shown in Figure 2, the higher construct existed, which is deemed as EC, and every fitting index of CFA as shown in Table 6. According to the judgment standard of the model fit,^36,59 the acceptable scope are GFI > 0.9, adjusted goodness-of-fit index (AGFI) > 0.9, and RMSEA < 0.1. The results in Table 4 are close to the requirement; therefore, hypothesis 1 is verified.

Figure 2.

The first-order and second-order confirmatory factor analysis of the emergency collaboration questionnaire.

Table 6.

Results of confirmatory factor analysis for emergency collaboration.

Model	χ²/df	GFI	AGFI	CFI	IFI	TLI	RMSEA
First-order model	2.933	0.811	0.744	0.752	0.758	0.701	0.111
Second-order model	2.892	0.810	0.747	0.753	0.759	0.708	0.109

CFI: comparative fit index; AGFI: adjusted goodness-of-fit index; GFI: goodness-of-fit index; IFI: incremental fit index; TLI: Tucker–Lewis index; RMSEA: root mean square error of approximation.

Factor analysis of quick emergency response capacity

The KMO and Bartlett’s test (Table 7) show that the KMO measure is 0.806, and the significance level of Bartlett’s test is less than 0.001. Therefore, the sample of ERC is suitable for factor analysis. Three common factors are extracted; the cumulative variance can explain 62.97%, the table of factor loading is shown in Table 8, and the factor loading less than 0.4 has been deleted, and the bigger one is reserved.

Table 7.

Kaiser–Meyer–Olkin and Bartlett's test of Emergency Response Capability.

Kaiser–Meyer–Olkin measure of sampling adequacy	0.806
Bartlett’s test of sphericity
Approx. chi-square	597.576
df	55
Sig.	0.000

Table 8.

Rotated component matrix of emergency response capability.

Item		1	2	3
VAR18	The emergency is observed as soon as passible	0.648
VAR19	The professional rescue team get there quickly	0.743
VAR20	The emergency materials are prepared rapidly	0.779
VAR21	The emergency equipment can be quickly implemented	0.793
VAR22	The leader of the emergency rescue plays an important role in the organization and coordination		0.540
VAR23	The commander plays a guiding role to conduct the emergency plan		0.636
VAR24	The commander plays a coordination role to allocate the emergency resources		0.772
VAR25	The leader can make the right decision using existing information		0.821
VAR26	Intra-organization can achieve emergency coordination			0.718
VAR27	Inter-organization can cope with the emergency together			0.765
VAR28	The organizations engaged in emergency rescue share the information in a high level			0.829

The first-order and second-order CFA are conducted through the real surveyed specimen (n = 159). As shown in Figure 3, the higher construct existed, which is deemed as quick ERC, and every fitting index of CFA as shown in Table 9. According to the above judgment standard of the model fit, the results in Table 4 are close to the requirement; therefore, hypothesis 2 is verified.

Figure 3.

The first-order and second-order confirmatory factor analysis of emergency response capability questionnaire.

Table 9.

Results of confirmatory factor analysis for emergency response capability.

Model	χ²/df	GFI	AGFI	CFI	IFI	TLI	RMSEA
First-order model	2.902	0.879	0.805	0.861	0.864	0.813	0.110
Second-order model	2.902	0.879	0.805	0.861	0.864	0.813	0.110

CFI: comparative fit index; AGFI: adjusted goodness-of-fit index; GFI: goodness-of-fit index; IFI: incremental fit index; TLI: Tucker–Lewis index; RMSEA: root mean square error of approximation.

SEM

Hypothesis 3 illustrates the relationship between EC and quick ERC. Taking the EC as exogenous latent variable, and the quick ERC or the capacity composition as endogenous latent variables, the SEM is established and calculated using AMOS version 22. The results are shown in Figures 4 and 5. The fitting indexes are shown in Table 10. Model I was developed first, with the latent dependent variable of EC and the latent dependent variable of quick ERC. According to the fitting effect, χ²/df is 2.371 < 5, RMSEA is 0.093 < 0.1, other indexes largely in line with the judgment standard. Therefore, the assumption is accepted and hypothesis 3a verified, which reveals that EC has a positive effect on ERC. In the next stage, Model II was developed by testing the path from EC to the three dimensions of ERC, respectively, were fast reaction and emergency disposal capability (FREDC), emergency decision and executive capability (EDEC), and coordination and joint action capability (CJAC). The second model resulted that χ²/df is 2.528 < 5, RMSEA is 0.098 < 0.1, other indexes largely in line with the judgment standard. Therefore, the assumption is accepted and hypothesis 3b verified, which reveals that EC has a positive effect on fast reaction and emergency disposal capability, emergency decision and execution capability, and coordination and joint action capability.

Figure 4.

The result of relationship validation between emergency collaboration and emergency response capability.

Figure 5.

The result of relationship validation between emergency collaboration and fast reaction and emergency disposal capability, emergency decision and execution capability, and coordination and joint action capability.

Table 10.

Results of structural equation model.

Model	χ²/df	GFI	AGFI	CFI	IFI	TLI	RMSEA
EC → ERC	2.371	0.749	0.702	0.711	0.717	0.680	0.093
EC → FREDC, EDEC, and CJAC	2.528	0.730	0.718	0.767	0.782	0.744	0.098

ERC: emergency response capability; EC: emergency collaboration; CFI: comparative fit index; AGFI: adjusted goodness-of-fit index; GFI: goodness-of-fit index; IFI: incremental fit index; TLI: Tucker–Lewis index; RMSEA: root mean square error of approximation.

Discussion and limitations

Discussion of results

Statistical analysis was carried out using SPSS version 22 and AMOS version 22; the reliability and validity of the questionnaires of emergency response collaboration and quick ERC were tested; factor analysis and path analysis were conducted for each variable; components of EC and quick ERC were discussed; interactions between EC and quick ERC were explored, and all of the hypotheses proved.

Dimensions of EC and quick ERC

Results indicate that EC includes four elements, that is, emergency organizations, emergency resources, emergency plans, and emergency information. Cooperating in emergency rescue, resources integration, implementing emergency plans, and information communication are necessary to realize inter-organizational EC. These four elements interact with each other. Talent, capital, materials, equipment, and other resources can promote organizational operation. Emergency plans arrange the issues of organizational operation, resource utilization, information transfer, and so on. Correspondingly, information delivered smoothly is pivotal to organizational operation and emergency plans implementation. Moreover, quick ERC includes fast reaction and emergency disposal capability, decision and execution capability, coordination and joint action capability. Quick ERC does not belong to any organization, but is shaped by activities and coordination in and inter-organization, that is, inter-organizational coordination ERC. Furthermore, the fast reaction and emergency disposal capability, emergency decision and execution capability, and coordination and joint action capability are indispensable and restricted by each other.

Relationship between EC and quick ERC

As we can see from Figure 4, EC has some influence on quick ERC (β = 0.28). Realizing inter-organization EC is conducive to improving ERC. The formation of quick ERC needs support from organizations, and resources, as well as implementing emergency plans and transmitting information smoothly. Further analysis, shown in Figure 5, indicates that EC has different impacts on different dimensions of emergency capability, that is, fast reaction and emergency disposal capability (β = 0.68), emergency decision and execution capability (β = 0.66), and coordination and joint action capability (β = 0.55). That is to say, the realizing of inter-organizational collaboration facilitates fast reaction and emergency disposal capability, emergency decision and execution capability, and coordination and joint action capability. Since different dimensions of quick ERC are closely related to each element of EC, allocating and controlling the elements, that is, emergency organization, emergency resources, emergency plans, and emergency information, reasonably and effectively, are key to realizing EC and generating quick ERC.

Other effects between EC and quick ERC

On the basis of the results of this study, EC can promote and foster the quick ERC in urban rail transit organizations. However, the effect of quick ERC on EC has not been validated in this study; moreover, whether there are any mediating and moderating variables can work on the relationships between EC and quick ERC. For example, emergency information can be a moderator which influences the efficiency of emergency response, or the different culture or atmosphere of different organization can enable or hinder the formation of EC. Hence, the relationships among many latent variables can be conducted and focused to facilitate the quick ERC.

Theoretical and managerial implications of results

Our research has important theoretical and managerial implications. Based on RBV and IPV perspective, information sharing and communication may be seen as complementary and intangible resources.¹⁷ Arguably, the tangible resources, that is, supplies, human resources, and plans, play important role in conjunction with the intangible resource of information sharing and communication. When combining or bundling those resources, they may lead to EC and/or ERC.^17,18 Our study provides empirical evidence that EC acts as an antecedent to quick emergency response capacity, and the content of EC also has prompting effect. This is one of the first studies utilizing survey data to test such hypothesized relationships in the urban rail transit context. ERC may be enhanced because better cooperation among organizations and resources allocating and information sharing can develop to promote emergency units responding more rapidly and effectively to disruptions.⁶⁰

In addition, this study offers some useful implications for emergency managers. Our findings demonstrate that investments in EC may generate ERC depending on key factors, such as adequate resources, good plans, open information channels, and so on. Thus, EC means avoiding operating in isolation. Using information, physical and other resources may get access to the beginning of the response. Our empirical results show that more willing the agencies are to share information, allocate resource, decision-making jointly, the better the response. Agencies should share information and develop partnerships to build trust long before an emergency occurs.¹⁵

Limitations and future research

The authors acknowledge that a sample size with larger coverage would be more appropriate for estimation using SEM analysis. Meanwhile, the sample size of this study was statistically adequate for developing a satisfactory SEM, because it is larger than the minimum requirement of 100 proposed by Hair et al.⁶¹ However, the sample of this study mainly focuses on the field of urban rail transit. Therefore, future research is recommended to conduct large-scale surveys covering professionals not only in urban rail transit, China, but also in other industries or western countries such as United Kingdom and the United states. The results of this study provide preliminary evidence and act as a platform for further studies that investigate the associations between EC and ERC across different industries and countries.

The respondents were all professionals who worked in urban rail transit administrations or firms and had direct experience with the emergency disposal or rescue, which should ensure the credibility of the data obtained in this study. Although the current quantitative study provides generalizable results of the relationships between EC and quick ERC, to enable an in-depth understanding of the relationships between these variables, qualitative research methods, such as case studies and focus groups, are recommended for use in subsequent research. Results from future qualitative data analysis may also act as cross-validation such as in the triangulation method.⁶²

SEM has become the methodology of choice for researcher investigating complex relationships between latent constructs.⁶³ Covariance-based structural equation model (CB-SEM), the most commonly used approach to SEM, is one of the confirmatory approaches that hold of the common factor to minimize the discrepancy between observed and estimated data through using software AMOS, LISREL, MPLUS, and so on. Our study also utilized CB-SEM method to test the relationship between EC and ERC. However, there is another SEM approach, namely variance-based SEM (partial least squares–structural equation model (PLS-SEM)), is one of the exploratory approaches that aimed to maximize the total variation of endogenous construct in a causal model. In comparison with CB-SEM, PLS-SEM requires less samples than CB-SEM,⁶⁴ is suitable for small sample and low response rates research. In contrast, we can try to adopt PLS-SEM method to conduct the emergency management research.

Footnotes

Acknowledgements

The affiliation for Zheng Junwei at the time this research was carried out is as above, however, he has now moved to Kunming University of Science and Technology, Kunming, China.

Academic Editor: Geert Wets

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by Ministry of Education in China (MOE) Project of Humanities and Social Sciences under Grant no. 13YJC630232 and Project of China Hunan Provincial science and technology Department under Grant no. 2014SK3214.

References

Dubey

Gunasekaran

The sustainable humanitarian supply chain design: agility, adaptability and alignment. Int J Logistic: Res Appl 2016; 19: 62–82.

Shen

Shaw

Managing coordination in emergency response systems with information technologies. In: Proceedings of the tenth Americas conference on information systems, New York, August 2004, pp.2110–2120.

Sun

. Research on comprehensive capability evaluation system of subway emergency based on process management. In: IEEE International Conference on Emergency Management and Management Sciences (ICEMMS), Beijing, August 2010, pp.500–503.

Beamon

BM.

Sustainability and the future of supply chain management. Oper Supply Chain Manage 2008; 1: 4–18.

Brandon-Jones

Squire

Autry

. A contingent resource-based perspective of supply chain resilience and robustness. J Supply Chain Manage 2014; 50: 55–73.

Dubey

Ali

Aital

. Mechanics of humanitarian supply chain agility and resilience and its empirical validation. Int J Serv Oper Manage 2014; 17: 367–384.

Gligor

Holcomb

Stank

TP.

A multidisciplinary approach to supply chain agility: conceptualization and scale development. J Business Logist 2013; 34: 94–108.

Wang

Liu

Evaluating emergency response capacity by fuzzy AHP and 2-tuple fuzzy linguistic approach. Expert Syst Appl 2012; 39: 6972–6981.

Loch

Terwiesch

Rush and be wrong or wait and be late? A model of information in collaborative processes. Prod Oper Manage 2005; 14: 331–343.

10.

Van Wassenhove

. Humanitarian aid logistics: supply chain management in high gear. J Oper Res Soc 2006; 57: 475–489.

11.

Kyng

Nielsen

Kristensen

. Challenges in designing interactive systems for emergency response. In: Proceedings of the 6th conference on Designing Interactive Systems, ACM, New York, 2006, pp.301–310.

12.

Thévenaz

Resodihardjo

SL.

All the best laid plans … conditions impeding proper emergency response. Int J Prod Econ 2010; 126: 7–21.

13.

Barney

JB.

Firm resources and sustained competitive advantage. J Manage 1991; 17: 99–120.

14.

Galbarith

JR.

Organization design: an information processing view. Interfaces 1974; 4: 28–36.

15.

Altay

Pal

Information diffusion among agents: implications for humanitarian operations. Prod Oper Manage 2014; 23: 1015–1027.

16.

Yeniyurt

Kim

. The impact of information technology on supply chain capabilities and firm performance: a resource-based view. Indus Market Manage 2006; 35: 493–504.

17.

Sirmon

Hitt

Ireland

RD.

Managing firm resources in dynamic environments to create value: looking inside the black box. Acad Manage 2007; 32: 273–292.

18.

Grant

RM.

The resource-based theory of competitive advantage: implications for strategy formulation. Calif Manage Rev 1991; 33: 114–135.

19.

Sirmon

Gove

Hitt

MA.

Resource management in dyadic competitive rivalry: the effects of resource bundling and deployment. Acad Manage 2008; 51: 919–935.

20.

Galbraith

. Designing organizations: an executive briefing on strategy, structure, and process. San Francisco, CA: Jossey-Bass, 1995.

21.

Schweitzer

Zimmermann

Mühlenbein

Coordination of decisions in a spatial agent model. Physica A: Stat Mech Appl 2002; 303: 189–216.

22.

Duncan

What is the right organization structure? Decision tree analysis provides the answer. Organization Dyn 1980; 7: 59–80.

23.

Simatupang

Sridharan

The collaborative supply chain. Int J Logist Manage 2002; 13: 15–30.

24.

Mileti

. Disasters by design: a reassessment of natural Hazards in the United States. US: Joseph Henry Press, 1999.

25.

McEntire

DA.

Reflecting on the weaknesses of the international community during the IDNDR: some implications for research and its application. Disaster Prevent Manage 1997; 6: 221–223.

26.

McEntire

DA.

Coordinating multi-organizational responses to disaster: lessons from the March 28, 2000, Fort Worth tornado. Disaster Prevent Manage 2002; 11: 369–379.

27.

Waugh

Jr.

Terrorism, homeland security and the national emergency management network. Public Organization Rev 2003; 3: 373–385.

28.

Comfort

Zagorecki

Coordination in rapidly evolving disaster response systems the role of information. Am Behav Sci 2004; 48: 295–313.

29.

Kapucu

Collaborative emergency management: better community organising, better public preparedness and response. Disasters 2008; 32: 239–262.

30.

McGuire

Silvia

The effect of problem severity, managerial and organizational capacity, and agency structure on intergovernmental collaboration: evidence from local emergency management. Public Admin Rev 2010; 70: 279–288.

31.

Wang

Improving emergency response collaboration and resource allocation by task network mapping and analysis. Safe Sci 2014; 70: 9–18.

32.

Robinson

Berrett

Stone

The development of collaboration of response to Hurricane Katrina in the Dallas area. Public Work Manage Policy 2006; 10: 315–327.

33.

Chiu

Zheng

Real-time mobilization decisions for multi-priority emergency response resources and evacuation groups: model formulation and solution. Transport Res Part E: Logist Transport Rev 2007; 43: 710–736.

34.

USNR Commission. The accident at Three Mile Island fact sheet. Washington, DC: US Nuclear Regulatory, 2000.

35.

Lindel

Prater

. Development of a decision support system for evacuation management. The first International Conference on Debris-Flow Disaster Mitigation Strategy, 2002.

36.

Post

Jernigan

Southeast United States Hurricane Evacuation Traffic Study: behavioral analysis, technical memorandum, Tallahassee, 2000.

37.

Llavador

Letelier

Penadés

. Precise yet flexible specification of emergency resolution procedures. In: Van de Walle

Turoff

(ed.) Proceedings of the 3rd international ISCRAM conference, Newark, NJ, May 2006, pp.110–120.

38.

Zarboutis

Marmaras

. Emergency management respecting system’s complexity: the case of fire in a metro railway system. In: Bagnara

. (eds) Cognition, culture & design, proceedings of ECCE-11, Catania, Italy, 8–11 September 2002, pp. 275–282.

39.

Bruyelle

O’Neill

El-Koursi

. Improving the resilience of metro vehicle and passengers for an effective emergency response to terrorist attacks. Safe Sci 2014; 62: 37–45.

40.

Mohammadfam

Bastani

Esaghi

. Evaluation of coordination of emergency response team through the social network analysis. Case study: oil and gas refinery. Safe Health Work 2014; 6: 30–34.

41.

UNDP. Capacity assessment and development: in a systems and strategic management context. Technical advisory paper, no. 3, bureau for development policy, 1998.

42.

Rao

Sun

A preliminary framework to measure public health emergency response capacity. J Public Health 2006; 14: 43–47.

43.

Chou

Chang

A decision support system for supplier selection based on a strategy-aligned fuzzy SMART approach. Expert Syst Appl 2008; 34: 2241–2253.

44.

Liu

Xie

LL.

Application of analytic hierarchy process in evaluation of cities’ ability for earthquake disaster prevention and reducing. J Nat Disast 2008; 17: 48–52.

45.

Tian

Yang

Study on the evaluation index system model of the emergency capability on emergency. J Basic Sci Eng 2008; 16: 200–208.

46.

Cui

GD.

Research on the evaluation of subway disaster emergency response capability based on rough-set theory. J Inst Disast Prevent 2010; 12: 13–19.

47.

Malhotra

Grover

An assessment of survey research in POM: from constructs to theory. J Oper Manage 1998; 16: 407–425.

48.

Churchill

GA.

A paradigm for developing better measures of marketing constructs. J Market Res 1979; 16: 64–73.

49.

O’Leary-Kelly

Vokurka

RJ.

The empirical assessment of construct validity. J Oper Manage 1998; 16: 387–405.

50.

Flynn

Schroeder

Sakakibara

A framework for quality management research and an associated measurement instrument. J Oper Manage 1994; 11: 339–366.

51.

Loehlin

Latent variable models: an introduction to factor, path, and structural analysis . 3rd ed. Mahwah, NJ: Lawrence Erlbaum Associates, 1998.

52.

Malhotra

Kim

Patil

Common method variance in IS research: a comparison of alternative approaches and a reanalysis of past research. Manage Sci 2006; 52: 1865–1883.

53.

Lambert

Harrington

Measuring non-response bias in customer service mail surveys. J Business Logist 1990; 11: 5–25.

54.

Armstrong

Overton

TS.

Estimating nonresponse bias in mail surveys. J Market Res 1977; 14: 396–402.

55.

Bollen

KA.

Structural equations with latent variables. New York: John Wiley & Sons, 1989.

56.

Norman

Draper

Applied regression analysis. New York: John Wiley & Sons, 1998.

57.

Landau

Everitt

. A handbook of statistical analysis using SPSS. London: Chapman & Hall, 2004.

58.

Liu

Kuang

Gong

. Principal component regression analysis with SPSS. Comput Meth Progr Biomed 2003; 71: 141–147.

59.

Kaplan

The impact of specification error on the estimation, testing, and improvement of structural equation models. Multivar Behav Res 1988; 23: 69–86.

60.

Jüttner

Maklan

Supply chain resilience in the global financial crisis: an empirical study. Supply Chain Manage 2011; 16: 246–259.

61.

Hair

Black

Babin

BJ.

Multivariate data analysis. Upper Saddle River, NJ: Pearson Prentice Hall, 2006.

62.

Cooper

Schindler

PS.

Business research methods. New York: McGraw-Hill/Irwin, 2010.

63.

Astrachan

Patel

Wanzenried

A comparative study of CB-SEM and PLS-SEM for theory development in family firm research. J Fam Business Strategy 2014; 5: 116–128.

64.

Ringle

Sarstedt

Hair

JF.

Partial least squares structural equation modeling: rigorous applications, better results and higher acceptance. J Long Range Plan 2013; 48: 1–12.