1

Alert Based Disaster Notification and Resource Allocation

Dickson K.W. ChiuSenior Member, IEEE

kwchiu@acm.org, dicksonchiu@ieee.org

Drake T. T. Lin , Eleanna Kafeza, Minhong Wang,

Haiyang Hu, Hua Hu, Yi Zhuang

Disaster Management 2

Introduction Disaster increasing frequency and severity

epidemic outbreaks, natural disasters, major accidents, terrorist attacks …

SARS, Avian flu, tsunami, … IS for disaster management badly required Unified framework for supporting all phases and activities

different prediction, detection, and specific handling activities variety of organizations and personnel heterogeneous physical and information resources dynamic process and information integration approach human and system interactions resource allocation under urgency constraints timeliness in action

Disaster Notification and Resource Allocation System (DNRAS)

Based on Alert Management System (AMS)

Example: epidemic outbreaks

SARS, Avian flu, … Retrieve and analyze related patient records

Monitor symptoms and patterns of spreading Records by fax or e-Mail result in mis-

interpretation Serious consequences Timely notification - nearby countries Resource allocation

Medicines and equipment Medical professionals

Many parties in the interactions and negotiations Many estimations and uncertainties

Disaster Management 3

Disaster Management 4

Requirements Overview

Both human and computerized systems are involved Stakeholders have different degree of computerization Web Services supports both type of interactions in a unified

framework

Broadcast /Internet Media

NotificationInformationReporters

Pharmacies /HealthcareSupplies

Internet & mobile devices

Local Government Authorities

Storage

Notification information

Sales

Storage

Administrators

Emergency Services

Hospitals

Storage

Physician

Physician

Public/Victims

Experts Officials AdministratorsPhysicians Mobile Individuals

Disaster Management 5

Role of Alerts in Strategic IS / DSS

Disaster Management

Processes and Workflows

DNRAS Actions:

Notification, Information

Request, Resource

Allocation

Alerts Managed by AMS

Web Services and Mobile Devices

What are Alerts? Different from general events,

alerts have more specific attributes, e.g., urgency and process requirements.

Different from exceptions, they need not relate to abnormal behaviors.

asynchronously received through business events / exceptions / incoming requests

synchronously generated by internal business application

handled by the AMS by requesting services from:

internal information systems management / human attention external e-Service providers

3R – Retry, Reroute, Reassign

Disaster Management 6

System Architecture

Database

DNRAS Scheduler Application

Logic

Web / WAP Access

Web Service Server

XSLT ProcessorWeb Portal

Web Services Programmatic

Access

Public UDDIRegistry

Triggered Action

Alert Input

XSLT Stylesheets

Desktop Laptop PDA Mobile

Public /Victims

Experts Officials AdministratorsDNRAS of Government,

Hospitals, Media, Supplies, Emergency Services

Internet

Disaster Notification & Resource Allocation System (DNRAS) Node

Alert Management System (AMS)

Service Provider Monitor

Urgencies Strategy Definition

Role Matching

Outgoing Alert Monitor

Process Alert Definition

Process Execution

Incoming Alert Monitor

AdminPersonnel /Call Center

Administrative Interface

Physicians

InternalInformation

Systems

Disaster Management 7

Alert Life Cycle

System flow of a DNRAS node

Disaster Management 8

DNRAS SchedulerApplication Logic

Receive request alertfrom other DNRA node

Generate and queuerequired request

Return an update response tothe request node

Find the required informationin cached database

Information complete

Aggregate the results

Update the informationto cached database

Send acknowledgementResponse

Fetch next queuing request

Update response Information request

Send request alertto next node

Information not complete

Reschedule the request

Alert ManagementSystem

Disaster Management 9

Alert Urgency Elevation Defining the policies according to which the

urgency of the alert will be “elevated” Example

dtdtdtTdtdtT CriticalVery

dtdtTdtT Critical

dtTTt Very Urgen

(default) T Urgent

)(002

32121

211

1

t

t

t

t

tU

Urgency002 Action

Urgent Default Very Urgent Submit a second alert to the pharmacy, notifying it about the

approaching deadline

Critical Redirect the alert to another pharmacy or hospital that has the best expected response time

Very Critical Send the alert to several pharmacies and hospitals and accept those that respond early; notify an administrator

Disaster Management 10

Web Services Summary

Notification alerts - attend to the verification and identification of the specific disease problem using medical information

General information alerts - inquires for general information Resource alerts - identify the place that can provide the

requested information or resource Personal information alerts - give information for a specific

person. Database query / update Web services. See paper for detailed parameter examples.

Advantages of Web services System integration - flexible process orchestration in disaster

management is able to integrate with legacy applications. Communication and Monitoring - the DARNS infrastructure

provides an interoperable system architecture for creating an efficient communication and monitoring infrastructure in order to respond to disasters efficiently.

Intelligence - Complex disaster problems can be identified and diagnosed in a timely way by alert-based data communication and information aggregation, and can be handled through flexible resource allocation and process orchestration.

Scalability - easy to extend the system by adding more web service based functionalities into each node of DNRAS to improve data communication and process coordination.

Reusability - The proposed architecture and Web services are reusable by other applications for flexible process orchestration in disaster management.

Disaster Management 11

Lessons Learned

Studied Toronto SARS outbreak 2003 case. Scarborough-Grace hospital were not alerted to the

possibility of an epidemic when several relatives of the first victim were found to be suffering from related symptoms and just considered TB

Delays in identifying the outbreak are vital. Direct and efficient communication among different agencies

required. When number of cases aggregated from the hospitals of a city

reached a specific threshold, an alert for an epidemic disease could be propagated to the public health office of the region.

Lack of isolation rooms – ICQ quarantined for 12–14 days. But continue to transfer patients to other hospitals!! Effective communication and resource allocation is critical

Disaster Management 12

Lessons Learned (2)

Physicians require efficient and immediate access to personal information files related to an epidemic outbreak.

Second SARS victim died in York Central Hospital had been next to the first victim in Scarborough-Grace hospital.

Information requests should be categorized structured, monitored, and associated with temporal deadlines.

Efficient information retrieval, aggregation, and association might lead to an early diagnosis of the disease

Communications, exchange of information, and notifications are crucial.

Disaster Management 13

Applicability Discussion Government authorities

Resolve existing problems involved in unreliable manual procedures

Handle exceptions Liaise with variety of parties and personnel AMS monitors / tracks such alerts Trials and simulations

Emergency service providers (hospitals / pharmacies) Fast rescue and recovery Locate resource holders for quick delivery Decision coordination

Mobile individuals (healthcare professionals) Ensure necessary and correct personnel involved Records / information retrieval Improved communications

Disaster Management 14

Disaster Management 15

Summary DNRAS, which supports alert notification and resource

allocation in the event of a disaster Cached information in the node and nested query requests

to aggregate information AMS for the coordination of various functions at various

stages of a disaster outbreak, including detection, notification, remedy, and recovery.

DNRAS can be built and plugged into the existing infrastructure of various stakeholders to bridge internal systems and external partners to form a grid

Improved information and process management

Disaster Management 16

Future Work

Privacy control of personal records Exploring various settings of the urgency tables Context-awareness in ubiquitous communication

management Complexity involved in the communication processes user communication management with agents Failure of commitments and their relation to contract

enforcement Impact of cancellations, other possible exceptions E-negotiation subsystem for negotiating costs and allocation System dependability, such as redundant connection links

and nodes

Disaster Management 17

Q&A

Thank you!