1 Introduction

Natural disasters such as pandemic outbreaks, floods, earthquakes, landslides and many more can strike in so many ways at any time. Meanwhile, a human-made disaster occurs due to development, human activities that include acts of terrorism such as bombing and actions to sabotage and contribute to the malfunction of critical national infrastructures such as electricity, gas and water facilities. United Nations Sustainable Development Goals outlined the most critical items during a disaster are water, food, fuel, emergency kits, first aid, and medical supplies [1]. Thus, the availability and access to the essential items to all areas especially those affected by disaster must be highly prioritised. Disaster can result in disruption of transportation and communication that lead to confusion on the standard operating procedure during the disaster event. The natural disaster that affecting the world is the Pandemic Covid-19 outbreak that caused severe illness, including death around the globe and creating a new phenomenon such as lockdown and restriction in terms of people mobility due to the disaster. The global outbreak of Covid-19, a cluster of acute respiratory illnesses with unknown causes, which started in Wuhan, Hubei Province, China, in December 2019, had made us experience the true nature of the seriousness of disaster caused by virus pandemic. Besides providing treatments and developing vaccines, tremendous efforts are put in place to detect and break the virus's chain, including introducing movement restriction orders around the world. To control the further spread of the Covid-19 pandemic, the Malaysian government has implemented various types of Movement Control Order (MCO). However, various levels of understanding by the authority members and people who received the instruction that keeps changing due to the various level of seriousness of the event caused a different compliance level within the community and varying levels of enforcement by the authorities. This gap can be solved by the utilization of technologies to assist the coordination of disaster management and release effort when the disaster structs. Development of mechanisms that apply smart city-related technology, telematics and big data analytics can be utilised in executing the strategy for disaster management and prevention.

This study proposes the conceptual model of personalised smart mobility by integrating the smart city concept for smart movement during a disaster that is crucially needed for efficient coordination and disaster relief effort. The model considers related policies as the foundation and components of relevant factors in the decision-making process in any possible event for the mobility of vehicle and people in essential sectors when disaster strike and the people movement is limited or restricted. This paper is organised as follows: first, we briefly discuss the background and general overview of existing approaches in handling disaster management focusing on Malaysia’s disaster management strategy that includes natural disasters such as pandemics, floods and landslides. This is followed by a discussion of the technologies that can be applied to manage disasters better. Then, we discuss the utilisation of data, information, and communication technologies (ICT) that can help to manage the pandemic in organised and coordinated ways. Finally, we propose a Personalised Smart Mobility Model for Smart Movement during a disaster by considering the smart city concept for urban Malaysia to ensure that people are informed with the updated situation, standard operating procedure (SOP) in real-time and advising citizens to comply with the authority instruction efficiently.

2 Background

The Malaysian government had coordinated all relevant government authorities and agencies, including public healthcare, police, defence, fire departments, welfare departments and so on, in handling the disaster including the Covid-19 pandemic disaster. Systematic and secured data sharing among all relevant government authorities and agencies is essential to enable efficient data analysis to support various decision-making, action planning, and implementation to preserve the nation's stability and security during a disaster. The knowledge domain that could be utilised in disaster management and coordination is the smart city, telematics and big data analytics using various sensors including smartphones. A disaster management strategy was also included as a pillar in the implementation strategy so that it could be applied efficiently with the assistance of technology.

2.1 Disaster Management Strategy

When disaster strike, to be able to respond to varying situations regardless of the scale of effects, decision support for disaster management requires flexibility and efficiency [3]. Involvement of expertise from different fields is essential as response involves multiple agencies and coordinating disaster management. According to the locality of the area, the precise management support system will also play a vital part when the disaster occurred [4]. In Malaysia, the National Disaster Management Agency (NADMA) handled disaster management and response under Directive 20 of the National Security Council. NADMA acts as the focal point in managing disaster with three (3) levels of disaster management that included district level, state level and national level. Each group has specific agencies that will be deployed for the relief operation during the disaster. Any disaster involving local incidents that can potentially spread will be managed by the District Disaster Management and Relief Committee (DMRC). If the disaster level escalated and the next level of support is required, the State Disaster Management and Relief Committee (SDMR) will manage the disaster. On the highest level, the Central Disaster Management and Relief Committee (CDMRC) will be in charge and responsible for forming a central, state and district level of operation team [5].

In flood and landslide scenarios, people mobility is also very significant to be handled carefully to ensure smooth evacuation of people from affected areas. It can be done by dynamically monitoring the disaster affected areas through cameras, smartphones and other IoT technologies to avoid road congestion, blockages and chaos [6, 7]. The study by [8] proposed a novel emergency response framework that facilitates the stakeholders of the urban transport systems in the decision-making during emergencies by emphasising on personalised mobility profile for evacuation during a disaster. Meanwhile, during the pandemic Covid-19, Malaysia National Security Council (NSC) commanded and decided on mobilising the pandemic response at the national level, with technical guidance from the Ministry of Health (MOH). Covid-19 was deemed a national security threat, the National Security Council (NSC) was activated via the National Security Council Act 2016 (Act 776) to command and mobilise all government and non-governmental machinery to control the pandemic on a national scale. Legislation and policy documents that applied during the pandemic disaster are the Prevention and Control of Infectious Diseases Act (Act 342), International Health Regulations 2005, the MOH Disaster Management Plan, and the Malaysia Strategy for Emerging Diseases and Public Health Emergencies (MySED) II Workplan (2017–2021) [9].

Also, the government made a recommendation to society to adhere to the SOP developed by the Ministry of Health, the National Security Council and the Prime Minister's Department. Preventive efforts to oversee individual and neighbourhood wellbeing was emphasized. Government officials are being advised to prevent Malaysia’s population from misinterpreting signals or directives issued by the authorities. Municipal authorities’ collaboration with the local government was established to manage Covid-19 in the community. Community councils also giving the best effort in keeping the community secure by collaborating with age-specific organisations to reach out to people in need [10].

However, for this scale of the disaster, the management of the disaster is overwhelming the responding agencies. In all aspects of disaster management, technologies that supported by various type of data and tools have great potential to expedite the coordination, decision making, save lives, limit damage and reduce the costs.

2.2 Smart City

The proliferation of the Internet of Things (IoT) devices, notably in the advancement of sensors technology, leads to the development of a Smart City worldwide. One example of such Smart City planning and development in Malaysia is the City Brain project in Kuala Lumpur [11, 12]. Smart City elements are comprised of citizens (people), transportation, healthcare, technology, governance, infrastructure, building, energy and education. Leading technologies in Smart City are sensors, cameras, mobile phones, high-speed wireless Internet and emerging new technologies that are part of the IoT. These devices generate big data for various utilisation purposes in Smart City. Big data refers to high volume, variety, velocity, and veracity [13].

2.3 Telematics for Vehicle and Citizen

Telematics refers to the transmission of information provided by sensors from various areas of the smart city using telecommunications services. Meanwhile, applications and services in the smart city can be divided into sensors, local processing, visualisation and background processing and analytics [14]. The implementations will include computational infrastructure, network facilities and data centres. The increase in worldwide smartphone penetration has gained new ways to collect data. This phenomenon resulted in benefitting people, vehicle owners, industry, government and society. A combination of sensors and smartphone usage is determined to be the best for data collection and communication sharing. A variety of connected devices and smartphones enable report writing and status update to be done in real-time. Also, based on data analysis, commands will be sent to the specific group resulted in an appropriate action that could be done [15, 16]. The mobility of smartphones triggers various research in smartphone-based vehicle telematics because it enables services based on the monitoring of devices and their surroundings. Among the research areas that need to be addressed are efficient methods for real-time decision-making. Also, smartphone usage gives the advantage in terms of application development that are optimising the user experience, human factors and improving the total cost of utilisation. A personalised mobility route planning model for campus evacuation was proposed in [17, 18]. The model includes registration functions and collecting the students’ information, providing the distributed location information for general emergency evacuation in the university campus, and smart personalised students’ route planning in an emergency evacuation.

3 Personalised Smart Mobility for Vehicles and People During Disaster

Based on the background studies in the previous chapter, we propose personalising citizens and vehicles for smart mobility using a unique profiling mechanism. The unique profile is created for a particular user based on their assigned role.

3.1 Supporting Technologies

The IoT devices that are embedded in the connected vehicle and installed throughout the roads and public transportation facilities for Smart City, can be exploited as telematics solutions to personalise the smart mobility profile. In a way, it provides accurate real-time smart mobility information of the vehicle, such as location, identification, capacity available and situation of the traffic. The current smartphone application that uses Global Positioning System (GPS) is handy for normal daily lives, such as knowing the route and traffic condition when travelling using cars. But in a disaster situation, GPS alone will not provide the true nature of the traffic situation. Additional IoT devices in Smart City such as Laser Detection and Ranging (LiDAR) [19], cameras and information from social media must be pooled as the input information to make a prediction of the routes. Through the information generated, authorities will have the predictive information on the fastest, safest and most efficient route [2, 20] to deliver the critical items supply to the eagerly waiting disaster victims. Using the Smart City concept, big real-time data of people and traffic mobility, for example, real-time GPS, traffic and road conditions data that are generated from the connected vehicles, various sensors and devices can be obtained and utilised in the process of labelling area according to the level of severity of the disaster. The collected data can be analysed using identified components, variables and parameters for telematics solutions to manage traffic flow and help with navigation strategies to provide emergency services. It could also be used for smooth evacuation from the affected areas, avoid congestion, minimise risk to public security and safety and minimise economic losses.

3.2 Model Conceptualisation

In this paper, we present the personalised smart mobility in the smart city for smart movement during a disaster by extending the Malaysia Smart City Framework. Our proposed model consists of 3 modules which are 1) Data obtainment module 2) Decision-making for vehicle and people mobility using big data analytics techniques module 3) Role-based profiling module as described in Fig. 1.

Fig. 1.
figure 1

Personalised smart mobility model for smart movement during disaster: Malaysia case

Full size image

Based on our conceptual studies, we determine the components and variables for the proposed model based on Malaysia Smart City elements focusing on monitoring and decision making for enforcement control, coordination and assistance arrangement during a disaster. We suggested the personalised profile be allocated to people (citizens) and vehicles (people who drive the designated vehicles) using the role-based profiling concept to enable optimum movement and mobility during the disaster. Meanwhile, the data gathered using telematics and smartphones will be used for decision-making to monitor traffic, vehicles and people's movement in real-time. This would help the authorities in implementing assistance arrangements and mobility controls during a disaster through personalised smart mobility tailored by each user's profile.

This model also takes into consideration parameters such as area zoning according to severity, safety and security definition that must be parallel with the government policy for disaster management tailored specifically for particular events. As for the collected data, this proposed model also comprised of the data privacy perimeters to protect the Confidentiality, Integrity and Availability (CIA) of the data. We discuss the module in the sub-section as follows:

3.2.1 Module 1: Telematics in Smart City for Dataset Obtainment

In this module, the data obtainment mechanism using telematics devices in and near the smart vehicles is designed to obtain data from vehicles to accelerate data processing and data feed structured with pre-defined variables. The vehicle telematics concept is utilised to organise systematic vehicles mobility using vehicles movements data across a specific area in a smart city. Then, real-time data analysis will be executed and the finest mobility movement according to each user movement objective, will be communicated using smartphones directly to each user.

The process used for data obtainment plays an essential role in mobility controls and assistance arrangement management. The method of generating model diagrams for the data obtainment for the analysis purposes be addressed by considering the stakeholders(actor) stance because the current understanding of the detail's activities, sequence, and decisions is still not precisely defined. Data from sensors, including connected cars, LiDAR, camera and video for the smart city mapping and traffic control, will be gathered in cloud computing, edge computing and fog computing based on the processing requirement. Then, the big data generated from the sensors and devices will be analysed according to the role of the data. Users such as citizens and vehicles, especially for those with crucial functions such as ambulances, need to have their personalised profile created and be integrated into the system.

3.2.2 Module 2: Decision-Making for Vehicle and People Mobility Using Big Data Analytics Techniques

Module 2 discuss the decision-making mechanism for vehicles and people mobility using data analytic. The decision-making variables will include vehicles and people specific profiles for precise data input. Based on the input data, the decision-making mechanism for particular circumstances by utilising the data analytics technique is proposed. We are proposing traditional data analytic techniques including Naïve Bayes, Recurrent Neural Network and Long Short-Term Memory to be utilised. The data that will be used are gathered in module 1 which include data from user’s smartphones, road data, event data, traffic sensors data and weather data. Under each component, variables are proposed as shown in Table 1. The decision-making is based on Malaysia's MCO scenarios when vehicles and citizens with essential roles determined by the government, need to access information about the best and safe routes to deliver food supply, clean water and healthcare supply in all areas. Meanwhile, in the case when citizens are having a symptom of the pandemic, this module will be able to guide them on the best routes to reach the medical team without risking other people's health by having unnecessary contact. This module also takes into consideration, the heavy traffic problem that occurred during the MCO due to excessive and unnecessary movement. It also will reduce the citizens’ non-compliance behaviour by having passive monitoring of people movement behaviour based on the collected data and will be sending the warning when citizens breach the SOP attached to their role. We also include the attributes such as the road network context-awareness data as a component for decision-making. We set the context as events, weather and structure affection. The features also include modelling pedestrian profiles, visual clustering of users’ mobility behaviour and route planning. Route planning consists of the evacuation and collaborative rescue procedure. This model also works in the evacuation during the disaster such as flood, landslide, fire, terrorism and any situation deemed as a disaster by the authority.

Table 1. Decision-making element for vehicles and people mobility
Full size table

3.2.3 Module 3: Role-Based Profiling

When a disaster such as a flood and a landslide occurred, the movement of vehicles that carries basic essential items to the various areas during a disaster must be prioritized. The availability and access for people and vehicles carrying essential needs to the designated location must be guaranteed. On top of that, for a pandemic disaster such as Covid-19 that require movement control to be implemented, the mechanism for people and vehicle mobility need to be systematically developed.

Role-based profiling modules are suggested in our model to assign a role to citizens (including authority personnel) and vehicles. This module will enable vehicles assigned with a particular role such as ambulance, police, army, transporting critical items etc. to have a safe and smooth route during the disaster using their personalised smart mobility profile. The authorities will be able to guide them to the best possible routes for delivering the much-needed critical items to the respected facilities or victims. As an application during MCO, low-risk citizens in a designated place that is considered as a low-risk area based on the profiling will be able to perform activities categorized as allowed activities based on the SOP by employing our proposed model. The components and variables that we proposed for this module are shown in Table 2. Under each element, variables are proposed. For example, for component authorities, the variables include police, fire brigade, military and so on as per definition by the government.

Table 2. Components and variables: role-based profiling for transportation
Full size table

Meanwhile, to increase the decision-making precision by utilizing data that was gathered by module 1, components to make profiling for a particular user based on their activities in the society must be defined when assigning the role. Base on the components, the specific variables that address user-specific roles for this module are shown in Table 3. Under each component, variables are proposed. The components and variables can be in concurrence with the present interpretation of the policy and situation.

Table 3. Components and Variables: Role-based profiling for People (Citizen)
Full size table

In a conclusion, in our proposed model, people personalised mobility information will be determined based on the elements, components and variables that set up for disaster management. To allow efficient role assignment for each citizen’s and vehicle’s personalise profile, they will be able to request their roles to the authorities and the approval will be given according to the recent SOP and policies set up by the government. Then, based on assigned role, the citizens and vehicles movement either public transport, personal transport or commercial transport during a disaster will be systematically managed. In a disaster, the dimensions, instruments, and parameters will be changing. Thus, for efficient and fast decision making, data analytics is utilised to analyse the data and provide instruction in accordance with the policies set by the government.

All 3 modules proposed for this model is designed to manage citizens and vehicles mobility according to their roles to give more flexibility and also authorization. The conceptual model is based on the disaster management strategy, literature review and case study during the events of disaster in Malaysia. As a result, this model will be contributing to systematic management and coordination during a disaster in real-time.

4 Conclusion

This research's main contribution is the elaborated concept and modelling of smart profiles for vehicles based on data gathering using telematics. The telematics ecosystem will contribute to big data generation. Thus, decision-making during a disaster that requires analysis can be done using big data analytics techniques to handle a large volume of data by taking into account all required parameters, components and variables. The application of telematics solutions for personalised smart mobility models using IoT and big data analytics in disaster management is an emerging and interdisciplinary research area that calls for collaborations between researchers with various backgrounds, government agencies and industries.

This article presented a Personalised Smart Mobility Model for Smart Movement during Disaster by applying the smart city concept in urban Malaysia. With this proposed personalised mobility profile, smart mobility can be implemented during the disaster and pandemic such as Covid-19. Even if a measure to counter the disaster such as MCO is executed, this proposed concept will help to categorise people with risk levels according to contagious patterns and map the highly affected area. The vaccination factor also must be added to the personalised profile to measure the risk of the citizen spreading the pandemic or being infected by the disease. As a result, only people with the risk of being infected will need to be quarantined until they are free from the disease. Also, geographically, only partial areas must be labelled as high risk and that area must be managed under the quarantined and restriction order precisely. Other than that, economic activity, education and learning, welfare and wellbeing, religious activity and celebration should be allowed to work as usual. This model will minimise risks and consequences related to disaster and should be included in the Malaysia disaster preparedness items [21]. As future work, an empirical analysis will be carried out by prototyping and establishing the simulation of the proposed model in the testing environment.