It is generally maintained that learning should be a part of the daily routines of many organizations; this is often referred to as lesson learned processes. The purpose of organizational learning is to foster improvements that seek to both reduce incidents and accidents and reduce their consequences when they nevertheless happen. Safety work is widespread among many organizations, e.g. aviation, hospitals, process industry, fire departments and several armed forces. A considerable part of safety work involves accident prevention, and aims to investigate why and how previous accidents and incidents happened, in order to learn how to avoid them, or minimize losses when they do occur. The collection of information after incidents represents one of the first steps in a lessons learned process, and the result is crucial for further work. Unfortunately, incident reports often tend to be unfocused (they represent a very wide area of issues) and, for that reason, cannot be clustered. They also frequently lack by analysts required information. The overall research objective in this thesis was to develop a report structure that enables the individuals who participated in or observed an incident to provide more information that is relevant about that incident. The first research question seeks to identify whether the Swedish Armed Forces face the kinds of problems that have been identified in earlier research on attempts to learn from accidents and incidents. The second and third research questions aim to ascertain whether the scope and quality of collected information in incident reports can be improved and if the number of incident reports can be increased. The results agree with earlier research and show that many of the problems that are common in other organizations (e.g. aviation, hospitals and the process industry) can also be observed and are a reality within the SwAF. In addition, the results showed that both scope and quality of collected information can be influenced. Group reporting using a consensus process neither had an appreciable effect on the quality of collected information, nor on the quantity of the reports. On the other hand, the new reporting form, which was based on interview and questionnaire methodology, and to some extent witness psychology, significantly improved the quality of the information collected after incidents. The new form proved to be superior, regardless of the character and context of the incidents. The information collected was also in accordance with what had actually happened and, finally, the form proved to be useful when various military “real world” incidents were reported. Finally, the results also provide new insights into the problems and possibilities associated with acquiring useful incident reports. The problem seems not only to be that people may be unwilling to report incidents that they have participated in or witnessed; it is also that they may be unable to do so. Consequently, it may not be sufficient to change the culture of the organization into a learning culture to receive by analysts required information. It is also necessary to help people report what they actually know by means of an improved report structure.
Huge amounts of data generated on social media during emergency situations is regarded as a trove of critical information. The use of supervised machine learning techniques in the early stages of a crisis is challenged by the lack of labeled data for that event. Furthermore, supervised models trained on labeled data from a prior crisis may not produce accurate results, due to inherent crisis variations. To address these challenges, the authors propose a hybrid feature-instance-parameter adaptation approach based on matrix factorization, k-nearest neighbors, and self-training. The proposed feature-instance adaptation selects a subset of the source crisis data that is representative for the target crisis data. The selected labeled source data, together with unlabeled target data, are used to learn self-training domain adaptation classifiers for the target crisis. Experimental results have shown that overall the hybrid domain adaptation classifiers perform better than the supervised classifiers learned from the original source data.
One in every 70 people around the world is caught up in a crisis (natural disasters, conflict, climate change, etc.) and urgently needs humanitarian assistance and protection according to the OCHA. The humanitarian community assists millions of people every year based on emerging humanitarian needs. Most of the time, the conditions inside the countries, once the humanitarian needs data is collected, are not very conducive and required simple ways to collect data like paper-based data collection with simple questions. This data is later entered into a database or spreadsheet using rigorous and time-consuming data entry efforts. Dynamic changes in needs of people; numbers of partners involved; the complexity of evolving processes; and emerging technologies over time has led to a change in processes for data collection and management. This article is an attempt to capture humanitarian data collection best practices and the use of different technologies in managing data to facilitate humanitarian needs assessment processes for the Syria crisis.
Crisis response, including humanitarian operations, is a highly complex field and its effectiveness is challenged by the dynamic partnerships of organizations involved and critical field conditions. Serious gaming is recognized as an effective method for complex systems design and analysis. Given the criticality of complex humanitarian operations and the current challenges faced by humanitarians in crisis response, serious gaming could play an important role in this field. However, the full potential of serious gaming in humanitarian assistance has not been fully explored yet. This article examines the role of serious gaming in assisting humanitarian operations. A board game is developed and played to examine its role in facilitating requirement engineering and training for humanitarian missions. In the contribution, the authors show how they were able to address the vital challenges faced by humanitarian aid workers in crisis response. Additionally, the outcomes of game sessions and their implications for humanitarian operations of the future was discussed.
Human behaviour during crisis evacuations is social in nature. In particular, social attachment theory posits that proximity of familiar people, places, objects, etc., promotes calm and a feeling of safety, while their absence triggers panic or flight. In closely bonded groups such as families, members seek each other and evacuate as one. This makes attachment bonds necessary in the development of realistic models of mobility during crises. This article presents a review of evacuation behaviour, theories on social attachment, crisis mobility, and agent-based models. It was found that social attachment influences mobility in the different stages of evacuation (pre, during and post). Based on these findings, a multi-agent model of mobility during seismic crises (SOLACE) is being developed, and it is implemented using the belief, desire and intention (BDI) agent architecture.
To strengthen the capability of societies to manage severe events, it is vital to understand what constitutes crisis management capability and how this can be assessed. The objective of this article is to explore how interorganizational crisis management capability has been assessed in the scientific literature. A systematic literature review was performed, resulting in a dataset of 83 publications. A thematic analysis resulted in nine themes of crisis management capability being identified, where interaction was the largest one. Analyses resulted in a comprehensive overview of assessment methods within the themes. The evaluation methods were mainly applied on real cases rather than exercises. The present article contributes with an increased understanding of how crisis management capability is evaluated, as well as applicability and limitations of different methodological approaches. This insight is essential in order to conduct a valid assessment of crisis management capability and design exercises that increase this capability.
Terrorism targeting corporate bodies remains one the greatest risks to the most critical intangible asset of any organization: reputation. Thus, effective crisis communication is critical during and after terror crisis to mitigate further damage on the reputation. To date, many studies around the globe have tended to focus on the role of the traditional media during terror crisis, paying minimal attention to organisations' use of social media during terror crisis. Using a descriptive qualitative case study, this study examined the role of social media during 2013 Westgate Mall terror attack, in Nairobi, Kenya. Findings revealed that the Interior Ministry (IM) used Twitter as the preferred social media platform to communicate with various stakeholders. Accommodative crisis response strategies were the most used by the IM. However, the ministry was plagued with inaccuracies and inconsistencies in its responses on social media compromising reputation of the government further. Balancing the need for speedy response, accuracy and consistency, remained the greatest challenge for the IM.
Infectious diseases remain a threat to public health, requiring the coordinated action of many stakeholders. Little has been written about stakeholder participation and approaches to sharing information, in dynamic contexts and under time pressure as is the case for infectious disease outbreaks. Communicable-disease specialists fear that delays in implementing control measures may occur if stakeholders are not included in the outbreak-management process. Two case studies described in this article show how the needs of stakeholders may vary with time and that early sharing of information takes priority over shared decision-making. The stakeholders itemized their needs and potential contributions in order to arrive at the collective interest of outbreak management. For this, the results suggest the potential for improvement through development of “network governance” including the effective sharing of information in large networks with varying needs. Outbreaks in which conflicting perceptions may occur among the stakeholders require particular attention.
Research on technology-assisted crisis management have been centered on tools to assist the response phase of a disaster. Through a semantic network analysis of concept relations in the titles of the publications, the authors found that many theoretical tools exist for disaster management but have not been operationalized to take a holistic approach toward technology for resilience. There is a lack of emphasis on the design of information and communication technologies, besides geographical information systems, to assist other phases of the crisis management cycle, particularly the preparedness and mitigation phases for resilience. By operationalizing the MOVE framework in a case study, the authors discover factors critical to the design of informatics and visualization tools to support resilience. This study concludes with a conceptual design framework “Digital Crow's Nest” to show feasibility of technology design for disaster resilience analytics using open data sources.
Information systems (IS) in emergency management (EM) support situational awareness and agility during a disaster so that professionals do not only need to follow rigid pre-defined plans that might be unsuitable in the unfolding situation. To use IS effectively, managers need an understanding of the capabilities of these systems; this can be achieved through an appropriate set of educational courses. This article presents the results of the analysis of a survey that proposed EM and IS courses for master level programs. The survey was completed by 373 practitioners, academics and/or researchers with EM experience. All proposed courses were rated above a 4 on a 7-point scale for how essential they are to a curriculum. A qualitative analysis indicates that some low ratings were due to disagreement over the described course content. An unexpected finding was that a substantial number of respondents spontaneously expressed opposition to the use of IS for EM in general. Findings are discussed and a preliminary curriculum is proposed.
To strengthen the capability of societies to manage severe events, it is vital to understand what constitutes crisis management capability and how this can be assessed. The objective of this article is to explore how interorganizational crisis management capability has been assessed in the scientific literature. A systematic literature review was performed, resulting in a dataset of 83 publications. A thematic analysis resulted in nine themes of crisis management capability being identified, where interaction was the largest one. Analyses resulted in a comprehensive overview of assessment methods within the themes. The evaluation methods were mainly applied on real cases rather than exercises. The present article contributes with an increased understanding of how crisis management capability is evaluated, as well as applicability and limitations of different methodological approaches. This insight is essential in order to conduct a valid assessment of crisis management capability and design exercises that increase this capability.
After a risk has manifested itself and has led to an accident, valuable lessons can be learned to reduce the risk of a similar accident occurring again. This calls for accident analysis methods. In the past 20 years, a large number of accident analysis methods have been proposed and it is difficult to find the right method to apply in a specific circumstance. The authors conducted a review of the state of the art of accident analysis methods and models across domains. They classify the models using the well-known categorization into sequential, epidemiological, and systemic methods. The authors find that these classes have their own characteristics in terms of speed of application versus pay-off. For optimum risk reduction, methods that take organizational issues into account can add valuable information to the risk management process in an organization.
Understanding the current situation is critical in every natural disaster or crisis. Therefore, there is a need for accurate and up-to-date information about the scope, extent and impact of a disaster. The basis for this information is data that is available through a variety of sensors. Decision Support Systems (DSSs) support decision makers in disaster management, response, and recovery by providing early warnings, insights into the current situation and recommendations for mitigation actions. For this purpose, raw sensor data needs to be collected, analyzed, integrated, and its semantics need to be automatically understood by the system. This series of processes forms a generic sensor to decision chain. In this paper, we present solutions and technologies to integrate those steps seamlessly, also demonstrating how each step of the pipeline can be visualized.
Small and medium-sized enterprises (SMEs) represent 99% of enterprises in Germany and more than 95% in the European Union. Given the recent increase of natural disasters and man-made crises and emergencies, it seems an important economic goal to ascertain that SMEs are capable of maintaining their work, revenue and profit at an acceptable level. According to ISO 22301, business continuity management (BCM) is a holistic management process which identifies potential threats and their impact to an organization and serves as a framework to increase organizational resilience and response capabilities. Prior research identified that BCM is under-represented in SMEs and that their security level is partially in an uneconomical range. This article presents the analysis of interviews with 19 independent micro enterprises highlighting findings on their low crisis awareness, varying technical dependency, existing action strategies and communication strategies and proposing a categorization of micro enterprises as preventive technicians, data-intensive chains or pragmatic jumpers.
Logistics management is crucial for the effectiveness and efficiency of humanitarian operations. Performance measurement enables practitioners to identify improvement potentials and management capabilities with regards to their logistics management tasks. While many performance measurement approaches for humanitarian logistics exist in the scientific domain, its applicability in the practitioner communities is rather low. Main reasons for this mismatch can be seen in rather few ready-to-use concepts and supporting information systems. In this article, the design, development, and evaluation of an information system for a Balanced Scorecard for humanitarian logistics is presented. The approach is embedded in the design science research framework following the agile programming methodology. The main characteristics of the server client architecture are described and reflected through experiences from formative and summative evaluations. The results stress the importance of the applied design approaches and support the closing of gaps between information systems designers and humanitarian practitioners.
The explosion of user generated content in social media published from mobile devices has led to the concept known as “citizen sensing.” Although English has been adopted by many as a de facto standard international language, reports about events, such as disasters, are frequently provided by citizens in their local language in addition to English. Attempting to integrate citizen reports from many languages is a significant challenge. This article describes the tools that address this challenge to enable the support of citizen-sensing of landslide events reported worldwide. Multilingual support is based on the first unified cross-lingual dataset of word vectors for representing texts in multiple languages. The classification model based on the proposed cross-lingual word vectors outperforms the “native” and “translated” approaches based on monolingual word vectors. Furthermore, it does not require the creation of a separate training set in a local language or its translation to English.
Understanding the current situation is critical in every natural disaster or crisis. Therefore, there is a need for accurate and up-to-date information about the scope, extent and impact of a disaster. The basis for this information is data that is available through a variety of sensors. Decision Support Systems (DSSs) support decision makers in disaster management, response, and recovery by providing early warnings, insights into the current situation and recommendations for mitigation actions. For this purpose, raw sensor data needs to be collected, analyzed, integrated, and its semantics need to be automatically understood by the system. This series of processes forms a generic sensor to decision chain. In this paper, we present solutions and technologies to integrate those steps seamlessly, also demonstrating how each step of the pipeline can be visualized.
Domain adaptation methods have been introduced for auto-filtering disaster tweets to address the issue of lacking labeled data for an emerging disaster. In this article, the authors present and compare two simple, yet effective approaches for the task of classifying disaster-related tweets. The first approach leverages the unlabeled target disaster data to align the source disaster distribution to the target distribution, and, subsequently, learns a supervised classifier from the modified source data. The second approach uses the strategy of self-training to iteratively label the available unlabeled target data, and then builds a classifier as a weighted combination of source and target-specific classifiers. Experimental results using Naïve Bayes as the base classifier show that both approaches generally improve performance as compared to baseline. Overall, the self-training approach gives better results than the alignment-based approach. Furthermore, combining correlation alignment with self-training leads to better result, but the results of self-training are still better.
Local social media users share and access critical information before, during, and after emergencies. However, existing methods can identify local social media users only after an emergency has occurred, and only then discover a small proportion of users sharing information in a geographic area. To address these limitations, we introduce the method of Social Triangulation to identify local social media users who access community information before an emergency in order to develop emergency communications strategies that contribute to community resilience. Social Triangulation identifies local users vis-à-vis the community organizations they curate within their social networks and, as a result, helps reveal the information infrastructure of a community. Consequently, social triangulation can inform emergency communications planning by identifying “filter bubbles” among social media users loosely embedded in an information infrastructure, as well as community influencers who are well-positioned to redistribute official information during an emergency.
