An emergency response framework for improving crisis management in facilities: PLS-SEM approach Open Access

Sub-Saharan Africa’s emergency response infrastructure (ERI), particularly in disaster management and healthcare, requires improvement to effectively meet the needs of its rapidly growing population (Zhou et al., 2024; Adjei et al., 2025a; Abudu et al., 2025a). Modern societies depend heavily on complex infrastructure systems, including utilities, telecommunications and financial networks, which support economic activities and social stability (Dudenhoeffer et al., 2006). However, these infrastructures are vulnerable to both anthropogenic and natural hazards, making effective emergency response and rapid recovery essential for maintaining socioeconomic well-being. Resilient infrastructure enables communities to withstand disruptions and maintain stability during crises (Sharma and Gardoni, 2018). Governments, non-governmental organisations (NGOs) and community leaders, therefore, play a critical role in developing policies and systems that enhance preparedness and response to threats such as disasters and terrorist attacks (Adjei et al., 2025b).

ERI comprises interconnected systems of hardware, software, equipment, facilities, personnel, procedures and communication mechanisms that support coordinated emergency management (Sharma and Gardoni, 2018). These systems are essential for protecting lives, safeguarding assets and ensuring the continuity of critical services (Dudenhoeffer et al., 2006). Within the built environment, facilities such as commercial buildings, educational institutions, healthcare centres and public spaces often accommodate large populations, making them particularly vulnerable during emergencies.

Several incidents illustrate the risks associated with inadequate emergency response systems in facilities. For example, a stampede at a train station in Bihar, India, injured 58 people, while a nightclub stampede in Yaoundé, Cameroon, resulted in 16 deaths and several injuries (Ye et al., 2024). Similarly, China records approximately 200,000 fires annually, with about 30% occurring in buildings, posing severe risks such as burns, suffocation and smoke poisoning during evacuation (Wang et al., 2024). These events emphasise the need for effective emergency response mechanisms to ensure safety during crises like fires, earthquakes, floods or terrorist attacks (Adjei et al., 2025b). This study focuses on emergency response initiatives within facilities, highlighting the need for coordinated systems to manage risks, support evacuation and enhance disaster response.

Although previous studies have examined emergency response systems and disaster management practices (Abudu et al., 2025b; Matveev et al., 2021; Dudenhoeffer et al., 2006), many focus on individual components such as infrastructure resilience, evacuation planning or coordination mechanisms. Limited empirical research integrates both technical and non-technical factors influencing emergency response effectiveness within facilities. There is limited research using partial least squares structural equation modelling (PLS-SEM) to explore relationships in emergency response. PLS-SEM effectively analyses interconnected constructs and develops predictive frameworks. This study uses PLS-SEM to develop a framework for evaluating emergency response in facilities, integrating technical and non-technical aspects to enhance crisis management and disaster preparedness.

An emergency response system is a network of physical, organisational and technical systems coordinated to manage and mitigate emergencies, ensuring public safety, property rights and security (Wang et al., 2022; Dwarakanath et al., 2021; Gilmore and DuRant, 2021). Myeong et al. (2020) emphasise the importance of smart governance and stakeholder involvement in enhancing city services and democratic values. The Smartopia Centre in Gimpo utilises big data and the Internet of Things to enable proactive disaster safety, aligning with the Sendai Framework’s 2030 goal to improve multihazard early warning systems.

The Sendai Framework for Disaster Risk Reduction 2015–2030 outlines four priority areas to build resilience by preventing and reducing disaster risks (Bello et al., 2021). Effective disaster risk management (DRM) policies must address key risk dimensions, including vulnerability, capacity, exposure, hazard characteristics and environmental factors, to ensure a comprehensive pre-disaster assessment, prevention, mitigation, preparedness and response. These principles are also relevant to earthquake emergencies, where rapid damage assessment, early warning systems and coordinated evacuation and rescue operations are essential for minimising casualties and infrastructure losses.

Wang et al. (2022) developed a framework for assessing emergency response capacity, focusing on government spending, public safety, education, transportation and healthcare. Fekete et al. (2020) highlighted efforts in Iran to enhance disaster resilience through improved DRM and governance. Gupta et al. (2022) and Zualkernan et al. (2019) introduced an artificial intelligence, Internet of Things and cloud-based framework for emergency operations to enhance information dissemination and resource readiness. Bello et al. (2021) emphasised a multisectoral approach to DRM and resilient recovery. Integrating DRM into national planning is essential to developing better policies.

Sukhwani et al. (2019) developed a flood disaster management framework focusing on knowledge, technology and institutions. The knowledge aspect assesses flood-risk understanding among agencies and communities, while the technology aspect involves accurate monitoring and warning systems.

The study utilises Perera et al.’s (2020) emergency response framework, which includes technical components (risk knowledge, monitoring, forecasting) and non-technical components (warning dissemination, communication, preparedness, response capabilities). It employs systems and resilience theories to explore the interconnectedness of emergency response systems in facilities (Chen et al., 2017). Systems theory focuses on component interactions, and resilience theory addresses the anticipation and recovery from disruptions (Son et al., 2020).

Existing studies have developed frameworks for DRM and emergency response systems (Sukhwani et al., 2019; Wang et al., 2022; Bello et al., 2021). However, many focus mainly on technological systems or institutional governance, with limited integration of both technical and non-technical components at the facility level. This research highlights gaps in ERI, particularly in the non-technical aspects. Most frameworks also analyse emergency systems at national or city levels, with less attention to emergency response in individual facilities and the built environment.

Additionally, few studies have empirically validated integrated emergency response frameworks using quantitative modelling approaches such as PLS-SEM, despite the conceptual contributions of previous studies (Perera et al., 2020; Sukhwani et al., 2019; Abudu et al., 2026). This study, therefore, addresses this gap by developing and empirically testing an integrated ERI framework that combines technical and non-technical factors influencing emergency response effectiveness in facilities.

The proposed framework in Figure 1 addresses both technical and non-technical components of ERI. The framework was developed based on the constructs identified in the reviewed literature, as no single comprehensive model exists for assessing ERI in facilities. These constructs were organised conceptually to guide the empirical analysis using PLS-SEM.

Disaster risk is determined by the type of hazard, vulnerability, exposure and capacity. Effective management requires identifying hazards, understanding community vulnerabilities and utilising resources. A holistic approach is vital for planning and mitigation. Monitoring systems (Gilmore and DuRant, 2021; Abudu et al., 2025a), including street security and traffic surveillance (Sukhwani et al., 2019), use technologies to improve public safety and emergency response (Myeong et al., 2020; Khan et al., 2020). Community and religious involvement enhance preparedness and resilience (Damaševičius et al., 2023).

Advanced tools and methodologies are employed to predict disaster events (Basher, 2006), integrating meteorological (Damaševičius et al., 2023), geological (Kedia et al., 2022) and environmental data to provide timely alerts and improve preparedness (Nick et al., 2023). Disaster risk reduction (DRR) depends on comprehensive data analysis to identify patterns, develop strategies (Cvetković et al., 2021) and reduce future risks. Assessing disaster situations involves monitoring hazards, evaluating immediate effects and analysing real-time data. Social media and digital platforms support disaster preparedness through real-time communication, training and centralised databases (Sutton et al., 2024; Damaševičius et al., 2023). Effective DRR needs stakeholder cooperation, social and gender inclusion, clear policies, community involvement, infrastructure investment and education.

Emergency response capabilities rely on skilled personnel, advanced technology (Wang et al., 2022; Damaševičius et al., 2023), communication systems (Myeong et al., 2020) and coordinated planning. Effective responses include alert systems, evacuations and rescue operations using GPS and cameras (Fekete et al., 2020; Gupta et al., 2022; Bello et al., 2021). Post-disaster assessments enhance readiness by identifying gaps in disaster risk reduction strategies (Cho and Choi, 2024). Recovery involves damage assessment (Nick et al., 2023), rehabilitation (Mukhopadhyay, 2023) and community compensation (Mensah-Bonsu, 2022). Strengthening recovery requires addressing vulnerabilities and improving coordination (Cho and Choi, 2024). Documenting damage aids in resilient rebuilding (Sutton et al., 2024; Wu et al., 2022).

Resilient reconstruction integrates risk reduction, infrastructure strengthening and land-use improvements (Gupta et al., 2022; Wang et al., 2022). Effective recovery depends on cooperation among government, NGOs, private sectors and communities to ensure inclusive planning and resource allocation (Cvetković et al., 2021; Perera et al., 2020; Tanesab, 2020; Nick et al., 2023). Public investment in resilient infrastructure and community programs enhances long-term recovery and risk mitigation (Fekete et al., 2020; Bello et al., 2021). Adaptation strategies must evolve continuously, incorporating new technologies, updated policies and lessons from past disasters (Sutton et al., 2024; Than et al., 2020).

This study used a quantitative approach to assess and develop a framework for emergency response in Ghanaian facilities, enabling measurable data collection and analysis of variable relationships for evidence-based conclusions. A survey design with structured questionnaires was utilised to gather data from professionals, including facility managers, engineers, architects, quantity surveyors, building inspectors, and National Disaster Management Organisation (NADMO) officials, across 43 districts, municipalities and metropolitan assemblies in the Ashanti Region. The target population comprised 516 professionals involved in facility planning and disaster response. This number was determined from official records of facility management departments and NADMO offices across the study area, representing the full population of professionals for the study. The sample size for the study was 159, which is further justified for PLS-SEM analysis. According to Lu and Lu (2025), most PLS-SEM studies use a sample size of 100; however, larger sample sizes yield more accurate predictions. A pilot study with 30 participants from district assemblies and academia validated the data collection tools. Purposive and convenience sampling were used to select experienced professionals for the study.

Data were collected through a structured questionnaire divided into two sections. The first section gathered demographic information, including educational qualifications, professional roles, facility types and years of experience (see Table 1). The second section consisted of Likert-scale items (1 = strongly disagree to 5 = strongly agree) assessing factors influencing emergency response systems, covering technical aspects such as risk knowledge and monitoring (Table 2), non-technical factors such as warning dissemination and preparedness (Table 3) and outcomes of ERI (Table 4). After collection, the data were cleaned, coded and normalised using Min-Max scaling. Analysis was conducted using SPSS and AMOS, with descriptive statistics summarising respondent profiles. Mean values ranked the importance of variables measured on the 5-point Likert scale, reflecting average perceptions.

PLS-SEM was used to validate the framework and examine structural relationships, making it suitable for analysing complex models with multiple latent constructs in emergency response systems and crisis management. Goodness-of-fit was evaluated using root mean square error of approximation (RMSEA; ≤0.08), comparative fit index (CFI) and Tucker–Lewis index (TLI; ≥0.90), and Chi-square/df (≤3; Savalei and Bentler, 2006). Content validity was ensured through expert reviews, whereas construct validity was confirmed via factor analysis. Ethical standards were maintained through voluntary, anonymous participation and adherence to Institutional Review Board protocols, with clearance number IREC 287/24.

The Results are presented in two parts: first, respondents’ ratings of technical, non-technical and outcome factors (Tables 2–4); second, analysis of critical factors and their interrelationships informing the PLS-SEM model.

This section summarises the demographic profile of respondents, as shown in Table 1, including education, professional roles, facility types and work experience. Most respondents are well educated, with many holding bachelor’s degrees or Higher National Diplomas, and a significant number holding master’s degrees. Engineers represent the largest group (66.7%), followed by quantity surveyors, NADMO officials and building inspectors, whereas facilities managers and architects are underrepresented. This occupational imbalance can lead to sampling bias, underrepresenting facility managers and architects. While focusing on technical roles aids in studying emergency response systems, it limits the generalisability of findings across all groups in facility management and disaster response.

Out of 225 distributed questionnaires, 159 valid responses were retained for analysis after screening. The remaining were excluded due to not meeting the requirements for PLS-SEM analysis. Table 1 provides the background characteristics of these 159 usable responses, which constitute the final data set for the study. Respondents primarily work in commercial facilities (42.1%), with additional experience in religious, recreational and educational facilities, while healthcare and industrial facilities are less represented. Experience levels also vary across respondents, bringing diverse professional perspectives to the analysis.

This section presents the descriptive statistics of the technical factors influencing the effectiveness of emergency response in facilities. Mean scores and normalisation techniques were used to rank these factors, while standard deviation is presented only to indicate variability in responses, as shown in Table 2.

Table 2 highlights that knowledge, monitoring and forecasting factors are all important for effective DRR in facilities. Within knowledge of risk, understanding disaster dimensions, risk identification and environmental conditions emerged as the most critical areas. Monitoring factors, such as disaster and traffic monitoring, along with street security surveillance, were particularly influential. Forecasting factors emphasised the importance of data management and analysis of past and current disasters. Overall, 12 technical factors, including risk understanding, monitoring, communication networks, risk assessment and data management, were identified as critical (normalisation values ≥ 0.60) for strengthening ERI.

This section presents descriptive statistics for non-technical factors influencing emergency response effectiveness in facilities, using mean scores, standard deviations and normalisation to rank these factors (Table 3). Ranking within each sub-construct was based on normalised values, with ties resolved by prioritising lower standard deviations.

The analysis identified four key non-technical dimensions impacting emergency response: communication, preparedness, response capabilities and recovery. Critical factors included awareness of pre-disaster information, risk policy enforcement, post-disaster assessments and damage evaluation for resilient rebuilding. Thirteen critical factors highlighted the importance of effective communication, policy enforcement, capacity building, coordinated rescue operations and stakeholder collaboration to enhance Ghana’s ERI.

This section presents the descriptive statistics for the outcomes of ERI, including average scores, standard deviations and normalisation values used to rank the outcomes (Table 4). When indicators shared the same mean or normalisation value, rankings were determined based on the standard deviation, with lower deviations receiving higher ranks.

Respondents rated the outcomes of ERI highly, emphasising the importance of protecting lives and properties, enhancing the resilience of populations and assets and fostering collaboration between civil society and government. Other notable outcomes, including reducing vulnerability and ensuring coordination between national and local capacities, were also strongly supported. These findings underscore the vital role of integrated systems in achieving both immediate protection and long-term disaster management.

The measurement model assessment in the SEM was conducted in two stages. First, confirmatory factor analysis was performed using IBM SPSS AMOS 24 and EQS 6.2 to assess model fit and estimate key indices, as well as R-squared values. The second stage, based on Hair et al. (2016), applied PLS-SEM quality checks, including reliability tests (Cronbach’s alpha and composite reliability), indicator reliability (factor loadings), convergent validity (average variance extracted, AVE), and discriminant validity (Fornell-Larcker criterion).

The study reported on seven model fit indices, namely, χ²/df, CFI, incremental fit index (IFI), normed fit index (NFI), TLI, standardised root mean square residual (SRMR) and RMSEA, which were used to assess the model fit (adequacy). The normed chi-square recorded good fit (1.691 < 3) at the initial analysis. It was observed that the initial indices for CFI, IFI, NFI and TLI were 0.862, 0.861, 0.722 and 0.854, respectively. The indices fell below recommended thresholds, prompting model modifications to include correlated error terms. After correlation, the indices improved: CFI (0.918), IFI (0.919) and TLI (0.912) met the 0.90 threshold. The RMSEA recorded 0.051 < 0.08, indicating an acceptable model fit, and the 95% CI ranged from 0.046 to 0.056 within an acceptable range (<0.08). These results confirm that the measurement model achieved an acceptable fit for structural analysis. Correlated error terms were introduced based on theoretical relationships among similar items within the same latent constructs, such as monitoring disaster risk and communicating disaster information, as per Hair et al. (2016). This approach enhanced model fit indices while preserving construct validity.

This section evaluates the hypothesised relationships between key factors influencing emergency response in facilities and the overall emergency response outcome (OER). After validating the measurement model, the structural model was assessed using key indicators, coefficient of determination (R²), predictive relevance (Q²), path coefficients and the statistical significance of relationships to determine the model’s explanatory power and the strength of inter-construct influences as illustrated in Figure 2.

From the results in Table 5, the coefficient of determination (R²) for the factors influencing the effectiveness of emergency response was significantly high at 79.4% (R² = 0.794 and adjusted R² = 0.781). The Q-square value for the model was 0.782, indicating that the model’s predictive power was highly relevant.

The path analysis revealed that several factors significantly influence ERI. Risk knowledge showed the strongest effect (β = 0.637, Z = 8.847, p  < 0.05), as presented in Table 5. Other significant predictors were forecasting (β = 0.277, Z = 4.291, p  < 0.05), warning dissemination and communication (β = 0.245, Z = 5.131, p  < 0.05), preparedness (β = 0.280, Z = 7.482, p  < 0.05) and recovery capabilities (β = 0.300, Z = 4.616, p  < 0.05). These findings indicate that both technical and non-technical components contribute to strengthening ERI in facilities.

The hypothesis testing results show that H1 (Preparedness), H3 (Recovery capabilities), H4 (Warning dissemination and communication), H5 (Risk knowledge) and H7 (Forecasting) significantly influence ERI and therefore support the proposed framework. In contrast, H2 (Response capabilities) and H6 (Monitoring) were not statistically significant.

Path analysis showed that response capabilities and risk monitoring had limited effects on Ghana’s ERI. In contrast, preparedness, recovery capabilities, warning dissemination, communication, risk awareness and forecasting had strong positive impacts, emphasising their crucial roles in enhancing emergency performance. Understanding hazards and vulnerabilities is crucial for decision-making and operational readiness. Integrating localised risk data and socio-economic assessments into emergency planning enhances targeted mitigation and resource allocation (Sukhwani et al., 2019; Abudu et al., 2025a). Ongoing monitoring and evaluation in Disaster Management Systems are vital for predicting threats and coordinating timely responses (Khan et al., 2020).

Similarly, forecasts showed a statistically significant positive effect on ERI, with a path coefficient of 0.277, a z-value of 4.291 and a p-value of less than 0.05. Accurate forecasting with early warning systems and real-time data enhances emergency systems’ ability to anticipate and respond to disasters. Advanced technologies facilitate pre-emptive actions, like evacuations and resource mobilisation, reducing vulnerability and boosting resilience (Basher, 2006; Abudu et al., 2025b). The evolution of forecasting from theoretical modelling to real-time application has enhanced DRM, improving preparedness in emergencies. Effective forecasting relies on integrating geological and environmental data to provide timely alerts and support strategic planning (Khan et al., 2020; Damaševičius et al., 2023).

The findings shown in Table 5 indicate a statistically significant positive link between warning dissemination, communication and ERI, with a path coefficient of 0.245, a z-value of 5.131 and a p-value below 0.05. Timely communication and effective warning dissemination are crucial for emergency response. Reliable networks ensure critical information reaches stakeholders quickly, aiding coordination and reducing delays, highlighting the importance of structured communication in DRM (Cvetković et al., 2021). Effective communication involves sharing hazard information, educating the public, engaging stakeholders and maintaining updated databases. In Ghana, challenges such as limited public awareness and weak information flow can weaken early warnings (Mensah-Bonsu, 2022; Than et al., 2020). However, technological innovations, including automated alerts and social media, are enhancing emergency communication (Dwarakanath et al., 2021; Mukhopadhyay, 2023). Effective risk communication should be inclusive and context-sensitive, ensuring clear and credible messaging (Abudu et al., 2025c). Throughout the disaster cycle, strategies integrated into training, hazard databases and mobile alerts can enhance preparedness (Tanesab, 2020; Kedia et al., 2022; Damaševičius et al., 2023). Additionally, involving local religious and social organisations in awareness programs can strengthen community preparedness and resilience (Sutton et al., 2024).

Preparedness is linked to improved ERI, with a significant path coefficient of 0.280 (z-value: 7.482, p  < 0.05). Comprehensive preparedness enhances the effectiveness of emergency systems. Facilities that focus on planning, resource allocation and training respond better to disasters, reducing impact and recovery time. Poor preparedness can have serious consequences, as seen in Germany during the 2021 floods (Nick et al., 2023), and developing regions are particularly vulnerable due to inadequate structures and planning (Mensah-Bonsu, 2022). Essential elements such as hazard forecasting, early warning systems and risk assessments are crucial for timely responses (Cvetković et al., 2021). Preparedness efforts require both technical infrastructure, like shelters and logistics centres (Mukhopadhyay, 2023), and institutional capacity, including community engagement and risk communication (Perera et al., 2020). Challenges such as poor governance and low public awareness can hinder effectiveness. Inclusive engagement of government, civil society and marginalised groups is vital for effective disaster risk reduction (Nick et al., 2023; Perera et al., 2020). Promoting preparedness through education and community networks strengthens grassroots resilience (Ntim, 2023; Sutton et al., 2024). Additionally, aligning national DRR strategies with localised implementation protocols improves clarity and resource mobilisation. Investment in early warning systems, resilient infrastructure and localised mitigation remains fundamental to long-term preparedness (Bello et al., 2021; Damaševičius et al., 2023).

The study emphasises the strong connection between recovery capacity and effective emergency response initiatives (ERI). Key efforts, such as infrastructure repair and psychosocial support, are vital for restoring normalcy after disasters (Cho and Choi, 2024). Prompt damage assessments and legal disaster declarations are essential for recovery planning (Mensah-Bonsu, 2022; Cho and Choi, 2024). Pre-disaster planning guides rebuilding strategies, whereas accurate assessments enhance recovery (Fekete et al., 2020; Gupta et al., 2022). Investment in resilient infrastructure and adaptable recovery policies promotes sustainable disaster management (Cvetković et al., 2021; Sutton et al., 2024). The non-significant results for response capabilities (H2) and monitoring (H6) suggest these factors may not directly affect ERI in the studied area. Response capabilities are reactive and vary by institution, but monitoring systems may lack integration into emergency management frameworks. These findings highlight the need for improved institutional coordination, technological integration and capacity development to enhance ERI.

The study provides practical insights for facility managers and policymakers to strengthen ERI. Managers should prioritise risk knowledge, forecasting systems and effective communication platforms to improve preparedness and response capacity within facilities.

From a policy perspective, government agencies should enhance disaster risk reduction policies, early warning systems and stakeholder coordination. The proposed framework offers a practical guide for strategic planning, policy development and investment decisions to improve crisis management and resilience in facilities.

This study developed the factors influencing ERI in facilities using a PLS-SEM approach. The results demonstrate that both technical and non-technical components are critical for strengthening crisis management systems. The structural analysis shows that risk knowledge, forecasting, warning dissemination and communication, preparedness and recovery capabilities significantly influence ERI effectiveness. In contrast, response capabilities and monitoring were not statistically significant in the model. These findings highlight the importance of integrating disaster risk, predictive systems, policy enforcement, communication mechanisms and coordinated recovery planning to enhance facility-level emergency response.

Based on these results, the study proposes a framework for modernising ERI that combines technical elements, such as risk assessment, monitoring and forecasting, with non-technical dimensions, including institutional preparedness, communication systems and recovery planning. This integrated approach provides a structured, measurable basis for improving disaster preparedness, guiding policy development and supporting strategic investment in resilient infrastructure. The framework strengthens crisis management by enhancing institutions’ ability to protect lives, safeguard critical assets and improve coordination across stakeholders.

This study is limited to professionals within selected districts, municipalities and metropolitan assemblies in the Ashanti Region of Ghana, which may affect the generalisability of the findings to other contexts. In addition, the analysis relies on questionnaire data, which may introduce perception bias. Future research could expand the geographical scope, incorporate longitudinal data and explore the role of emerging technologies and community-based approaches in further improving ERI systems.