A Framework Linking Information Quality Dimensions and ...

Proceedings of the 42nd Hawaii International Conference on System Sciences - 2009

A Framework Linking Information Quality Dimensions and Coordination Challenges during Interagency Crisis Response Rafael A. Gonzalez and Nitesh Bharosa Delft University of Technology [email protected], [email protected]

Abstract Responding to large-scale crises requires coordination within and between multiple relief agencies. Such coordination depends on access to the right information at the right time, along with other information quality dimensions, such as completeness and consistency. Multiple evaluation reports of crisis response conclude that the coordination of agencies was poor due to information that was too late, incomplete or inaccurate. However, the relationship between coordination challenges and information quality dimensions remains ill-understood. The contribution of this paper lies in a conceptual framework that provides insight into the relationship between information quality dimensions and coordination challenges through information management activities. The framework is used to present the findings of a case study of crisis response exercises. Coordination challenges and information quality problems of the case are presented, together with information management activities as a source for improvement of the two.

1. Introduction Responding to large-scale crises requires the coordination of multiple response agencies, including police, medical and fire services. The presence of such agencies results in goal conflicts, cultural differences and competition over resources [1]. For example, one agency may be interested in immediate relief efforts, while another may be interested in long term reconstruction. From an informational perspective, their communication platforms might be incompatible, their language might differ, their access to and training in information systems will not be leveled, and common standards will be hard to create in advance. Besides requiring coordination, effective crisis response also requires access to the right information on the right level of detail at the right time. Such requirements depend on information quality (IQ)

dimensions, such as correctness, accuracy and timeliness. Multiple reports on crisis response have argued that lack of information quality is a problem in practice [2]. Our main premise is that effective coordination depends on high IQ across multiple information management activities. This means that having access to high quality information, and appropriately distributing or sharing this information between the agencies, improves the conditions for effective and efficient coordination. However, despite the multiple contributions on defining IQ (e.g., [3, 4]) we still do not fully understand the relationships between IQ dimensions and coordination, specifically with regards to the particular coordination challenges of crisis response. Furthermore, as discussed by Chen et al [5], coordination in the context of crisis response is an understudied issue. Accordingly, the main objective of this paper is to develop a framework for understanding the relationships between various IQ dimensions and coordination challenges through information management activities. This paper proceeds in Section 2 by briefly stating our research method. In Section 3 we introduce an understanding of coordination in the domain of crisis response and present a set of coordination challenges (CC). This is followed by presenting information management (IM) activities as the link between those challenges and IQ dimensions. Section 4 structures the three elements (CC, IM and IQ) into a framework where the claim is that IM activities, when coupled with IQ criteria, can contribute to dealing with CC. A case study, reported in Section 5, applies and shows the practical implications of the framework. Finally, discussion and conclusions are offered in Section 6.

2. Research method Current literature on interagency disaster response considers the issues of coordination challenges (e.g. [5, 6]), and information quality (e.g. [4]) separately (albeit, highlighting in some cases the importance of

978-0-7695-3450-3/09 $25.00 © 2009 IEEE

1


information management for coordination). In order to develop the framework for understanding the relationships between them, we conducted literature research on coordination challenges and IQ challenges during interagency crisis management. We searched for contributions on coordination and information quality in relation to crisis using the ACM Portal, Science Direct (Elsevier) and Synergy (Blackwell) databases. We concentrated on identifying definitions and dimensions or issues listed on both sides. As a result, we framed coordination challenges and information quality dimensions together through information management activities. The information quality attributes and the list of coordination challenges are not meant to be exhaustive or comprehensive, but rather suggestive of the kinds of issues and elements related to coordination and information quality. The link through information management activities addresses their relationship in the practice of crisis response. Consequently, we built a framework which claims that those information management activities, when carried out with information quality, constitute mechanisms for tackling coordination challenges. The framework is illustrated using a case study. Based on the components of the framework, we present the observations of a total of 6 crisis response exercises spread over three days, in which real relief agencies participated. The units of analysis during the exercises were the information manager and the Emergency Control Room (ECR) of the crisis response organization of the Port of Rotterdam. Inside the ECR we observed different stakeholders comprising multiple response disciplines and three hierarchical response levels (operational, command and regional levels). Using the framework as observation protocol, we registered events and issues of a qualitative nature and after the exercises triangulated these observations by comparing notes and reviewing crisis response documents from the Port of Rotterdam and the Rotterdam-Rijnmond Region.

3. Conceptual Background In this section we describe the conceptual building blocks of our framework: coordination, information management activities and information quality.

3.1. Coordination in Crisis Response During a crisis or emergency, many of the logistical problems are not caused by lack of resources, but by failure to coordinate their distribution [7]. Conversely, successful operations are often attributed to effective interagency coordination [8]. This makes coordination

in crisis response a critical issue for researchers and practitioners. A widely used definition of coordination was established by Malone and Crowston (1994) as managing dependencies between activities [9]. This definition is rooted in organizational theory, and especially in Galbraith´s information-processing view [10], which has been challenged for organizations where work is non-routine and highly contextualized; in this case, an alternate definition of coordination is: a temporally unfolding and contextualized process of input regulation and interaction articulation to realize a collective performance [11]. In the context of crisis response, several definitions of coordination have been offered, such as: • The degree to which there are adequate linkages among organizational parts (tasks and subunits) so that the organizational objectives can be accomplished [12]. • The act of working together harmoniously, which implies performing independent activities as part of a larger plan in which emergent conflicts are resolved through a coordination strategy [13]. Successful coordination, whether it focuses on dependencies or on collective goal achievement, requires effective communication and resource sharing; during a crisis, lack of resources and poor communication belong to the nature of an affected emergency area [6]. This means that coordination difficulties are endemic to crisis response, especially when it is an interagency effort. Some coordination challenges (CC) of interagency crisis response, according to Chen et al. [5], are: • (CC1) High uncertainty, sudden and unexpected events; • (CC2) Risk and possible mass casualty; • (CC3) Increased time pressure and urgency; • (CC4) Severe resource shortage; • (CC5) Large-scale impact and damage; • (CC6) Disruption of infrastructure support; • (CC7) Multi-authority and massive people involvement; • (CC8) Conflict of interest; • (CC9) High demand for timely information.

3.2. Information Management activities linking coordination challenges and information quality dimensions The previous coordination challenges are related to an information-intensive domain and dealing with them often depends on effective information

2


management, understood here as finding, integrating and sharing information. Achieving coordination between a disparate group of actors depends on their access to timely, valid information and their capacity for information search, exchange, absorption and adaptation [14]. However, this capacity is limited in humans: as information requirements increase, cognitive capacity decreases [10]. Accordingly, information management (IM) activities can be deployed to deal with this limitation. In this paper we consider three basic information management activities for crisis response [15-17]: • (IM1) Searching, collecting and gathering information about the incident. • (IM2) Collating, evaluating and analyzing the information gathered. • (IM3) Distributing, sharing and exchanging incident related information. These three activities need to be carried out within the complex and dynamic setting of crisis response. Such a setting imposes particular considerations for the information management activities to be successful. Some of these considerations are: • IM activities should be carried out both within and across agencies [16]. • IM activities need to be deployed within an information infrastructure that is both flexible and ordered [15]. • IM activities should focus on the core information of the incident, including intensity, location and related damages, as well as availability of resources [6]. • People and groups need to share interpretation schemas, or at least be able to make some conjectures on what other people's schema is [18]. • IM activities should be carried out as much as possible before the actual crisis occurs [19]. The main consideration that this paper focuses on is the need for the information management activities to be carried out with high information quality. As such, the IM activities become the bridge between information quality and coordination. This provides the basis for a framework aimed at improving information quality through information management activities that ultimately results in improved coordination of interagency crisis response.

3.3. Information Response

Quality

during

needs of relief workers [20] and poor IQ can be disastrous for both relief workers and victims [4]. Emergency managers have learnt that accurate and timely information is as crucial as is rapid and coherent coordination among the responding agencies [21]. Moreover, because disaster management agencies are information- intensive [22], their effectiveness largely depends on the information they have available [23]. IQ is a multidimensional concept, describing properties of the information received, capturing a wide range of variables such as accuracy, timeliness, completeness, consistency, relevance and fitness for use [24]. Miller [3] adds format, compatibility, security and appropriate amount of data as important variables for measuring the IQ. Table 1 lists some of the IQ dimensions and related problems pointed out in crisis response literature. Table 1. IQ problems pointed out in literature IQ Dimension (IQ1) Accuracy

In emergency management, information about technical conditions may be ambiguous and unreliable [25].

(IQ2) Timeliness

In analyzing an emergency, the situation changes over time so it is very important to know the order of events and their cause-effect relations [26].

(IQ3) Relevance

Certain events viewed in isolation may appear irrelevant or benign in terms of the emergency, but when analyzed collectively may identify a potential threat [27].

(IQ4) Quantity

When a large-scale disaster happens, a great deal of information occurs in a short period of time [26], resulting in too much information to process [28] and straining the capacity of the management and communication systems [29].

(IQ5) Completeness

There are potential delays influenced by the availability and dispatch of complete information about the incident [30].

(IQ6) Format

To enable information sharing, document type definitions have to be in a welldefined format. While the format of data is arbitrary, the format of data definitions needs to be rigorously defined [28].

(IQ7) Security

Sharing information during an emergency is challenged by trust and security issues [29], because of the need to protect potential misuse of information; however, excessive regulation hampers responders from getting useful information from other agencies [31].

(IQ8) Consistency

If several information systems suggest different location coordinates, this inconsistency delays decision making [4].

Crisis

During the process of interagency crisis response, information is considered to be one of the essential

IQ problem pointed out in literature

3


The above information quality dimensions, coupled with information management activities in order to deal with coordination challenges become the core of the framework presented in the next section.

4. Framework linking information quality dimensions and coordination challenges

In Table 2 we match the information management activities (IM1, IM2 and IM3) with the IQ dimensions (IQ1 to IQ8). We do this to present the potential benefits for coordination that can be achieved when carrying out each of the IM activities with IQ criteria. These benefits are at the intersections of IQ and IM in the table, based on our interpretation of the literature.

Table 2. A framework describing Information management activities and information quality dimensions related to coordination challenges IM Activities IQ Dimensions

(FR1) Search, collect and gather information about the incident.

(FR2) Collate, evaluate and analyze the information gathered.

(FR3) Distribute, share and exchange incident related information.

(IQ1) Accuracy

Finding accurate information is critical for assessing the potential risks of the incident (CC2).

Evaluating the information for accuracy contributes to better allocation of limited and distributed resources (CC4).

Exchanging accurate information about the incident contributes to determining the scale of the impact and damage (CC5).

(IQ2) Timeliness

Quick information gathering about the incident is critical in dealing with a dynamic environment and time-pressure (CC9).

Integrating and analyzing incident information with timerelated data, helps create a cause-effect picture of the event (CC1).

Distributing the information in a timely fashion to responders improves awareness of each others activities and decisions (CC9).

(IQ3) Relevance

Retrieving relevant information about the incident reduces the uncertainty (CC1).

Evaluating the relevance of a piece of information in relation to the incident, contributes to dealing with the speed and uncertainty of information (CC1).

Distributing information that is relevant to each different agency, contributes to faster decision-making (CC3).

(IQ4) Quantity

Collecting sufficient information about the incident from all agencies helps achieve common situation awareness and contributes to agreement on goals (CC8).

Filtering the information from multiple wide-spread sources helps prevent information overload (CC5).

Distributing the right amount of information to the right people helps deal with an increasing demand for information (CC9).

(IQ5) Completeness

Gathering complete resource availability information contributes to improved resource usage (CC4).

Evaluating for completeness contributes to better assessment of risks (CC2).

By distributing complete information about the incident, the consequences of communication breakdowns are reduced (CC6).

(IQ6) Format

The ability to gather information from different sources and in different formats allows for improved redundancy of communication channels (CC6).

Integrating information from different formats helps create a common picture and overcome technological differences between agencies (CC7).

Distributing information in the right format for different responders contributes to dealing with the heterogeneity of agencies (CC7).

(IQ7) Security

Enabling security in information retrieval prevents future conflicts of interest from occurring (CC8).

Analyzing security and privacy issues of incoming data helps determine priorities and distribution lists (CC7).

By preserving security, agencies will be more willing to share information (CC8).

(IQ8) Consistency

Collecting consistent information speeds up awareness and decision-making (CC3).

Evaluating information for consistency helps integrate the different sources from different agencies (CC7).

Ensuring consistency in information distributed across communication channels prevents conflicting decisions and action (CC8).

4


Table 2 embodies a framework linking information quality to coordination challenges through information management activities or services that can be implemented for crisis response. However, using the framework requires taking into consideration the relationship that exists between these activities and the potential tradeoffs between them. Firstly, IQ dimensions should be an integral part of the whole information management system used during a crisis response, but the complete system would include remote sensors, public sources of news, and ad hoc responders, among others, over which there might be little or no control. Secondly, the framework does not unveil potential tradeoffs involved in dealing with one information quality dimension over another. These tradeoffs need to be balanced and weighed against other information systems design principles for the framework to be effective. For example, some of the tradeoffs involved in the fifth row of Table 2 include: • Collecting complete information from all response agencies (IM1-IQ5) to contribute to a shared situational awareness of the incident may also impact the information-processing capacity, conducing to potential overload (IQ4); • Evaluating and analyzing information that is as complete as possible (IM2-IQ5) may improve the assessment of risks related to the incident, but it may also increase the time for decisions to be made and communicated (IQ2); and • Distributing complete information (IM3-IQ5) may reduce the impact of communication disruptions, but it may also require providing more information than is strictly relevant for a given agency or responder at a given time (IQ3).

Netherlands, where it contributes around 8% of its GDP and remains the largest in Europe [33]. Every day enormous quantities of chemicals and other hazardous materials are imported, transferred and stored at the PoR. Handling such substances implies potential catastrophes for humans and infrastructure in the area. In addition, large-scale emergencies or crises call for the participation of several regional agencies, which may include police, port authority, medical services, fire department, hazardous materials experts, and government officials. When an emergency occurs, a Regional Operational Base-Plan (ROB) is carried out to protect the physical and social health of the population [34]. The two decision making authorities for the ROB are the Command Place Incident (CoPI) fire, police, ambulance, hazmat specialists, and press and the Regional Operational Team (RegOT). Previous reports on studies in the PoR have shown several issues related to information management [35] and coordination [36]. In this paper, we focus on information quality and coordination challenges in PoR crisis response exercises, thus bridging those two previous reports. The goal of this present case study was to identify hurdles for information quality as potential sources for improving coordination during a crisis response, in order to illustrate the practical implications of the framework presented in the previous section. The crisis response exercises observed were organized indoors for three days, containing three rounds per day. The objective (as stated by the organizers) was formulated as: Introducing advanced information systems to relief workers, including commanders of the relief organization that participate on the decision making levels of disaster response. Table 3 provides an overview of the case study elements.

In the following section we present a case study that offers a setting for exploring the practical implications of our framework.

Table 3. Elements of the exercise Collective response to crisis using Focus

5 Case Study: Crisis Response Exercises at the Port of Rotterdam

Crisis management Levels

Operational level (first responders), tactical level (COPI), strategic level (ROT) and Emergency Control Center

We carried out a case study on interagency crisis response in the Port of Rotterdam (PoR). The purpose of the case study is instrumental as it is selected for its ability to demonstrate some phenomenon such as understanding of an issue or refinement of theory [32]. More specifically this case study illustrates some of the information quality problems and coordination challenges during interagency crisis management. The PoR is a mainport, a strategic link in a global supply chain that fuels the economy of The

Participating agencies

Police Department, Fire Department, Medical Services, DHMR (Division Harbor Master Rotterdam), CVD (Center for public service)

Setting

Computer aided simulation, Indoors

Schedule

June 5th, 19th and 26th 2007 (2 exercises each day)

ICT Tools

ICIS, CeDRIC, GROOVE

advanced ICT tools

5


The entire exercise was conducted in one room (without any sub rooms/compartments). In the center of the room a beamer/screen was set up, allowing for collective explanation and evaluation of the exercise. Figure 1 provides an illustration.

Figure 1. Schematic overview of exercise setup Sixteen workplaces circled the center of the room. The numbers at each workplace represent an agency participating in the exercise. Each workplace represented an actor and contained two PCs and a printer. Each actor was represented by one or two persons. The list of workplaces and the actor or role per agency present in the workplace can be found in Table 4. It is clear from those involved that the team inside the observed setting was the emergency control room, the CoPI and regional command, while operational actors were external (as would be the case during a real emergency). The emergency scenarios that were followed during the two rounds of exercises were basically the same, with the exception of three elements: the location, the magnitude and the information structure. The main incident of the scenario was the collision of two ships: a containership and a passenger ship. On the container ship an explosion occurs. Because of the potentially hazardous chemical material on the container ship there is a risk not only for the passenger ship, but also for the neighboring ships and factories in the busy setting of the Port of Rotterdam, thus the response requires multiple disciplines from professional emergency agencies, advised by domain and port experts.

Table 4. Workplaces in exercise Nr.

Emergency Service

1

Leader RegOT (Regional Operational Team)

2

Spokesperson CoPI (Commando Place Incident)

3

Information Manager CoPI

4

Chemical Advisor CoPI

5

DHMR (Division Harbor Master Rotterdam)

6

CvD (Center for Public Service)

7

Police Officers for CoPI

8

OD Fire Department

9

Leader CoPI

10

Spokesperson RegOT

11

Information Manager RegOT

12

Chemical Advisor RegOT

13

HS DHMR

14

HS Police

15

HS GHOR

16

HS Fire Department

Decision-making during crisis response in The Netherlands is defined in the Coordinated Regional Incident-Management Procedure (GRIP, according to the Dutch acronym Gecoördineerde Regionale Incidentbestrijdings Procedure). GRIP has four alarm levels, corresponding to increased levels of severity, and becomes active for incidents involving hazardous materials, large-scale technical emergency response, or any other accident where at least one of the CoPI members calls for a coordinated response to an incident [37]. Level 1 considers routine incidents requiring no collective decision-making between responding organizations. GRIP level 2 events entail small scale disasters requiring collaboration between multiple stakeholders. GRIP level 3 and 4 are considered large scale disasters in which the Mayors are required to take strategic decisions. Figure 2 illustrates the GRIP bottom-up structure as observed in the exercises. Figure 2 shows that there is a hierarchical relationship of command and control established from the bottom-up as the emergency escalates. When the response required is of a regional and multidisciplinary nature, the operational level is followed by a CoPI team and in the highest level a regional RegOT, all of which have representatives from the required disciplines and communicate through the backbone of the control room.

6


Table 5. Findings on the level of coordination challenges Coordination Challenge

Figure 2. Observed command structure Our main focus was the CoPI team where decision making is carried out in rounds of 15 minutes between the members, which are the operational leaders of the involved disciplines, led by a CoPI leader. The need for coordination and effective information management is critical for this team, which is why we also looked at the information manager, whose role it is to serve as liaison between the CoPI team and the different information sources available.

5.1. Coordination, information quality and information management in the case During our observations in the Port of Rotterdam, we focused on hurdles that challenged the level of information quality for responders at the command level, as well as on instances of general crisis response coordination challenges as observed during the exercises. Although some coordination challenges were not as strong, given the fact that they related to exercises, and not to a real emergency, they exerted a significant influence on the response and its associated information management. The observations, firstly, illustrate the framework with the case observations and, secondly, point at sources for improvement through information management services as discussed later in Section 6. Although the focus was on the command and leadership inside the exercise setting presented in the last section, information was flowing both towards and from the operational responders not physically present at the room. Improving coordination is one of the goals of crisis response exercises (e.g. through training and networking). But the exercises themselves also allow for detection of coordination challenges for the response organization. The level of the coordination challenges observed is rated on a three point scale (low, medium and high) and is shown in Table 5.

Level of coordination challenge observed

(CC1) High uncertainty, sudden and unexpected events

High: The scenarios of the exercises contained both the element of surprise and unpredictability, making in difficult to predefine information needs and coordination activities.

(CC2) Risk and possible mass casualty

High: risk in terms of uncertainty about the dangers was dominant in the first phases of the response.

(CC3) Increased time pressure and urgency

High: the time pressure was very high especially since the exercise time was short and the events occurred rapidly.

(CC4) Severe resource shortage

Low: the port area is known for having sufficient resources for handling at least one crisis at a time (this may be a problem in case of multi-event or distributed crisis)

(CC5) Largescale impact and damage

Medium: The scale of the disaster can be rated as medium because it was more a regional than a national issue.

(CC6) Disruption of infrastructure support

Low: in general there was no notable disruption of the infrastructure during the exercises (the exercises were conducted indoors). However, had it been a real emergency, any disruption of the PoR affects critical infrastructure.

(CC7) Multiauthority and massive people involvement

Low: As the observed case data is from an exercise, the number of agencies and decision making levels were limited.

(CC8) Conflict of interest

Low: despite their different expertise and work processes, the agencies were each committed to jointly reducing casualties and mitigating the effects.

(CC9) High demand for timely information

High: especially in the first minutes after the crisis is announced, there is a high demand for situational information coming from each agency.

In the current information architecture of the PoR there are multiple information managers, including a remote emergency control room, that have access to and share information with relief workers. Our focus was the CoPI information manager. As the disaster evolves and the complexity (in terms of amount of actors and necessary interactions) increases, we found more and more variations in the IQ. Because these variations are difficult to measure and quantify, we chose to report examples that illustrate the information quality level. Table 6 provides an overview of these

7


exemplary IQ issues, from the point of view of the information manager, for illustration purposes and not as an exhaustive list of information quality observations. Table 6. Findings on information quality issues IQ dimension

Observed IQ issues

(IQ1) Accuracy

In one exercise the initial coordinates of the ship collision were incorrect, leading to the concentration of field units in an area opposite to the real location.

(IQ2) Timeliness

The blueprint of one of the ships contained necessary information for firemen who had to go on one of the ships to mitigate the effects of the explosion. Getting access to the blueprint took about 20 minutes.

(IQ3) Relevance

Field units from the medical services were fed with information about climate conditions at the location, while they were far more interested in information about the victims.

(IQ4) Quantity

While incident information is scarce during the first 20 minutes of the exercise (low quantity), relief workers are overloaded with information from various sources with different quality values (high quantity).

(IQ5) Completeness

Information in the first 20 minutes of the crisis is incomplete during each exercise round, for instance about the number of victims on board.

(IQ6) Format

The blueprints of the ships are unavailable in digital format and need to be retrieved at a physical location.

(IQ7) Security

Access to information about potentially flammable materials stored in containers at the collision area is secured and not directly accessible.

(IQ8) Consistency

The members of the various agencies have different information on the exact location of the incident.

Having observed information quality issues and coordination challenges, the next step in using the framework presented in this paper would be to identify information management activities that could be carried out or improved across the different information quality dimensions, in order to deal with the coordination challenges. Table 7 presents our observation about the IM activities conducted at the two units of analysis: the

information manager and the Emergency Control Room. Table 7. Findings on information management Unit of analysis

Information Manager

Emergency Control Room

(IM1) Searching and collecting

Medium: the information manager was depended on the available tools which only allowed for the retrieval of Geo data and mailing. Low: Apart from the background of the Information manager (expert) there were no specific protocols for information quality assessment

Medium: The ECR basically only pushes incoming information from different sources to field units

Medium: The information manager only shared when asked by the head of the decision making unit

High: Filtered by the agency type represented in the ERC, information is distributed regularly to all relief workers.

(IM2) Collating, evaluating and analyzing

(IM3) Distributing, sharing and exchanging

Low: The focus of the ERC is primarily information push and pull without information quality assessment

Table 7 shows that both the Information Manager and The ECR primarily focus on sharing the information provided by the information systems they have access to. In this setting there is no predefined and exercised protocol for evaluating the quality of the information shared. Moreover, the current information systems used do not provide any services to support rapid IQ assessment. By supporting information management activities at the two units of analysis, considering the information quality issues (Table 6), an improved management of coordination challenges (Table 5) could be possible as we discuss in the next section.

6. Discussion and Conclusions This paper investigated the relationship between coordination challenges and information quality during interagency crisis response. This was motivated by the observation that in crisis response failure and success of coordination are often attributed to information quality issues. By reviewing literature, this paper listed some coordination challenges and information quality

8


dimensions. Information management activities were then used as the link between the two. This resulted in a framework which was applied in an instrumental case study in the Port of Rotterdam, both to find how empirical observations fit within the framework and what the practical implications are. We conclude that by designing computer-based information management services that embed information quality dimensions we can improve coordination of interagency crisis response. In the case of the Port of Rotterdam, introducing the use of these services would require the role of the information manager (IM). This is because introducing new technologies into crisis response often fails due to the highly specialized practices of responders. The information manager would fulfill the role of an intermediary in charge of information management activities. This mediation should initially play a prominent role, especially during training, to disseminate the new work practices and the use of the new information management services. On the other hand, the services themselves should be flexible enough to become part of an adaptive information architecture that can cater to responders from all the agencies as the role of the information manager gradually stabilizes and the information services become widely used. One way in which this can be achieved is by integrating coordination and information management through a workflow approach. Since workflows can coordinate complex processes where many activities must be scheduled and dispatched among many agents, one would like the workflow management system to offer access to information or to directly pump information into the appropriate place in the workflow [38]. The workflow approach can contribute to increased control, reduced response time, spotting of bottlenecks and rapid evaluation [19]. However, it has been recognized that, for workflow management, coordination across semantically heterogeneous environments (such as multi-agency crisis response) is a significant problem area, and that fixed workflows are inadequate for real-world situations (especially with the complexity and unexpected nature of a crisis) [39, 40]. Accordingly, any such effort must make use of adaptive or flexible workflows which may be based on goal-based reasoning [40], agent-based mechanisms that handle the coordination of the flexible workflow [39], or pre-existence of templates that can be embedded into the workflow in a case-based manner [19]. Implementing information management activities through workflows and additional roles, whilst maintaining information quality attributes, should conduce to improved coordination. Nonetheless, this

creates necessary but not sufficient conditions for such improvement, additional IQ attributes and coordination challenges could be added. In addition, we already mentioned in Section 3.2 some considerations for information management besides quality, and other moderating variables, such as overall information system quality and institutional structure, can also be subject of further research. This would enable our framework to be linked to previous, more general, design principles for crisis response information systems.

7. References [1] T. Bui, S. Cho, S. Sankaran, and M. Sovereign, "A Framework for Designing a Global Information Network for Multinational Humanitarian Assistance/Disaster Relief," Information Systems Frontiers, vol. 1, pp. 427442, 2000. [2] S. S. Dawes, A. M. Cresswell, and B. B. Cahan, "Learning from crisis: lessons in human an information infrastructure from the world trade center response," Social Science Computer Review, vol. 22, pp. 52-66, 2004. [3] H. Miller, "The multiple dimensions of information quality," Information Systems Management, vol. 13, pp. 79 - 82, 1996. [4] C. W. Fisher and B. R. Kingma, "Criticality of data quality as exemplified in two disasters," Information and Management, vol. 39, pp. 109-116, 2001. [5] R. Chen, R. Sharman, H. R. Rao, and S. J. Upadhyaya, "Coordination in emergency response management," Communications of the ACM, vol. 51, pp. 66-73, 2008. [6] L. K. Comfort, K. Ko, and A. Zagorecki, "Coordination in rapidly evolving disaster response systems: The role of information," American Behavioral Scientist, vol. 48, pp. 295-313, 2004. [7] T. C. Chan, J. Killeen, W. Griswold, and L. Lenert, "Information Technology and Emergency Medical Care during Disasters," Academic Emergency Medicine, vol. 11, pp. 1229-1236, 2004. [8] W. Smith and J. Dowell, "A case study of co-ordinative decision-making in disaster management," Ergonomics, vol. 43, pp. 1153-1166, 2000. [9] T. W. Malone and K. Crowston, "The interdisciplinary study of coordination," ACM Computing Surveys, vol. 26, pp. 87-119, 1994. [10] J. Galbraith, "Organization Design: An Information Processing View," Interfaces, vol. 4, pp. 28-36, 1974. [11] S. Faraj and Y. Xiao, "Coordination in fast-response organizations," Management Science, vol. 52, pp. 11551169, 2006. [12] R. R. Dynes and B. E. Aguirre, "Organizational adaptation to crises: Mechanisms of coordination and structural change," Disasters, vol. 3, pp. 71-74, 1979. [13] J. B. van Veelen, P. P. A. Storms, and C. J. van Aart, "Effective and Efficient Coordination Strategies for Agile Crisis Response Organizations," in 3rd International Conference on Information Systems for

9


Crisis Response (ISCRAM2006), B. van de Walle and M. Turoff, Eds. Newark, 2006, pp. 202-213. [14] L. K. Comfort and N. Kapucu, "Inter-organizational coordination in extreme events: The World Trade Center attacks, September 11, 2001," Natural Hazards, vol. 39, pp. 309-327, 2006. [15] L. K. Comfort, M. Dunn, D. Johnson, R. Skertich, and A. Zagorecki, "Coordination in complex systems: increasing efficiency in disaster mitigation and response," International Journal of Emergency Management, vol. 2, pp. 62-80, 2004. [16] R. Heath, "Dealing with the complete crisis-the crisis management shell structure," Safety Science, vol. 30, pp. 139-150, 1998. [17] M. Stephenson Jr, "Making humanitarian relief networks more effective: operational coordination, trust and sense making," Disasters, vol. 29, pp. 337-350, 2005. [18] R. Cuel and M. Cristiani, "Ontologies as IntraOrganizational Coordination Tools," in 5th International Conference on Knowledge Management (I-KNOW '05), T. K. and M. H., Eds. Graz, Austria, 2005. [19] T. Bui and A. Tan, "A Template-based Methodology for Large-Scale HA/DR involving Ephemeral Groups - A Workflow Perspective," in 40th Annual Hawaii International Conference on System Sciences (HICSS 2007) Waikoloa, HI: IEEE, 2007, pp. 34-34. [20] I. Helsloot, "Bordering on Reality: Findings on the Bonfire Crisis Management Simulation," Journal of Contingencies and Crisis Management, vol. 13, pp. 159169, 2005. [21] B. van de Walle and M. Turoff, "Emergency Response Information Systems: Emerging Trends and Technologies," Communications of the ACM, vol. 50, pp. 29-31, 2007. [22] H. de Bruijn, "One fight, one team: the 9/11 commission report on intelligence, fragmentation and information," Public Administration, vol. 84, pp. 267-287, 2006. [23] T. H. Davenport and L. Prusak, Working Knowledge, How Organizations Manage What They Know. Boston: Harvard Business School Press, 1998. [24] R. Y. Wang and D. M. Strong, "Beyond accuracy: what data quality means to data consumers," Journal of Management Information Systems, vol. 12, pp. 5-33, 1996. [25] T. Kontogiannis, "Stress and operator decision making in coping with emergencies," International Journal of Human-Computer Studies, vol. 45, pp. 75-104, 1996. [26] Y. Atoji, T. Koiso, and S. Nishida, "Information filtering for emergency management," in IEEE International Workshop on Robot and Human Communication Osaka: IEEE, 2000, pp. 96-100. [27] N. Adam, V. P. Janeja, A. V. Paliwal, B. Shafiq, C. Ulmer, V. Gersabeck, A. Hardy, C. Bornhoevd, and J. Schaper, "Approach for Discovering and Handling Crisis in a Service-Oriented Environment," in IEEE International Conference on Intelligence and Security Informatics (ISI 2007) New Brunswick, NJ, 2007, pp. 16-24

[28] J. Jenvald, M. Morin, and J. P. Kincaid, "A framework for web-based dissemination of models and lessons learned from emergency-response exercises and operations," International Journal of Emergency Management, vol. 1, pp. 82-94, 2001. [29] B. S. Manoj and A. Hubenko Baker, "Communication challenges in emergency response," Communications of the ACM, vol. 50, pp. 51-53, 2007. [30] R. Chen, R. Sharman, H. R. Rao, and S. Upadhyaya, "Design Principles of Coordinated Multi-incident Emergency Response Systems," in Intelligence and Security Informatics, P. Kantor, G. Muresan, F. Roberts, D. D. Zeng, F. Y. Wang, H. Chen, and R. C. Merkle, Eds. Berlin: Springer, 2005, pp. 81-98. [31] J. K. Kim, R. Sharman, H. R. Rao, and S. Upadhyaya, "Efficiency of critical incident management systems: Instrument development and validation," Decision Support Systems, vol. 44, pp. 235-250, 2007. [32] R. E. Stake, The Art of Case Study Research. Thousand Oaks, California: Sage Publications, 1995. [33] Statistics Netherlands, "Port of Rotterdam grows mainly through container transports," in Economic Monitor Voorburg: Statistics Netherlands, 2005. [34] J. Trijselaar, "Gecoördineerde Regionale Incidentenbestrijdingsprocedure Rotterdam-Rijnmond. Versie 4.2.," Rotterdam, 2006. [35] N. Bharosa, J. Appelman, and P. de Bruijn, "Integrating technology in crisis response using an information manager: first lessons learned from field exercises in the Port of Rotterdam," in 4th International Conference on Information Systems for Crisis Response (ISCRAM2007), B. van de Walle, P. Burghardt, and C. Nieuwenhuis, Eds. Delft, 2007. [36] R. A. Gonzalez, "Coordination and Its ICT support in Crisis Response: Confronting the InformationProcessing View of Coordination with a Case Study," in 41st Annual Hawaii International Conference on System Sciences (HICSS 2008) Big Island, Hawaii, 2008, p. 28. [37] D. J. Mendonça, G. E. G. Beroggi, and W. A. Wallace, "Decision support for improvisation during emergency response operations," International Journal of Emergency Management, vol. 1, pp. 30-38, 2001. [38] A. Abecker, A. Bernardi, H. Maus, M. Sintek, and C. Wenzel, "Information supply for business processes: coupling workflow with document analysis and information retrieval," Knowledge-Based Systems, vol. 13, pp. 271-284, 2000. [39] S. Wang, W. Shen, and Q. Hao, "An agent-based Web service workflow model for inter-enterprise collaboration," Expert Systems with Applications, vol. 31, pp. 787-799, 2006. [40] W. L. Kuechler Jr and V. K. Vaishnavi, "An expert system for dynamic re-coordination of distributed workflows," Expert Systems with Applications, vol. 34, pp. 551-563, 2008.

10