IEEE Technology and Society Magazine - Fall 2014 - 36

apid social, economic,
and political changes
are leading factors
that call for a shift in
the ways we respond
to continual events occurring daily.
Having a proactive response, rather
than a reactive response, allows decision makers to detect opportunities
and threats that could affect business
[6]. Eliminating or mitigating an
anticipated problem, or capitalizing
on a forecast opportunity, can substantially improve quality of life, and
prevent environmental and economic
damage. Changing traffic light policies and speed limits to avoid traffic
congestion, for example, will reduce
carbon emissions, optimize public
transportation, and improve the productivity of commuters. Similarly,
adding credit cards to watch lists
as a result of forecasting fraud will
reduce the cost inflicted by fraudulent activities on payment processing

made by forecasting events prior to
their occurrence. Motivation for proactive decision making stems from
social and economic factors, and is
based on the fact that prevention is
often more effective than the cure.
The decisions are made in real time
and require swift and immediate processing of Big Data, that is, extremely
large amounts of noisy data flooding
in from various locations, as well as
historical data.
Proactive applications have been
developed for several years [19], for
example, proactive security systems
and proactive routing in mobile ad
hoc networks. However, these applications have largely been developed
in an ad hoc manner. In contrast, in
this article we propose a new, generic
methodology for proactive eventdriven computing.
We are moving from the information economy to the "intelligent economy," where we have the capability

Real-time optimization of power
consumption is needed in individual
buildings equipped with renewable
energy sources.
companies and merchants, and consequently lower credit card rates.
In energy management, there is a
requirement for real-time optimization of power consumption in individual houses and buildings equipped
with renewable energy sources. This
requirement can be met by forecasting energy consumption and production, within the span of 30 minutes,
allowing for prompt decision making about load adjustments and/or
rescheduling within the a relatively
short time-frame. In post-earthquake
disaster management, loss forecasts
can be vital in planning the actions to
be taken immediately after an earthquake occurs.
To prevent problems and capitalize
on opportunities, we propose a methodology for proactive event-driven
decision making. Proper decisions are
36

not only to access information, but
also to analyze and act upon it. This
enables sustainable management of
communities, and promotes appropriate distribution of social, healthcare, and educational services [11].
To meet the shift to the "intelligent
economy," a methodology for proactive event-based decision making
is required. This methodology comprises the following steps. First, the
Big Data is continuously acquired
and aggregated from various types of
sensors. The aggregated data is analyzed and fused in order to recognize,
in real time, events and situations of
special significance. To allow for
timely recognition, communication
overheads must be minimized by
moving as little data as possible from
one place to another. Second, the recognized events are correlated with

historical information to forecast
problems and opportunities that
may actually take place in the near
future. Third, the forecast events,
along with the recognized events,
are leveraged for real-time operational decision making. Fourth,
visual analytics tools prioritize and
explain possible proactive actions,
enabling human operators to reach
and execute informed decisions.
Proactive computing requires
capabilities for forecasting, real-time
decision making, and visual analytics. These capabilities are extremely
important in a multitude of application domains. For example, currently
no system supports fraud forecasting.
Furthermore, a typical fraud detection system may recognize up to
nine false alarms for each true alarm.
Without an appropriate explanation
of why a specific transaction is considered fraudulent, the operator overseeing transactions will not be able
to confirm the fraud. They will either
have to let it go through, or contact
the end user, spending valuable time.
We propose exposing the user to how
proactive computing works through
visual analytics.
To summarize, the new methodology is designed to support onthe-fly, low-latency processing of
extremely large, geographically distributed, noisy event streams and historical data. The methodology will
recognize and forecast opportunities
and threats, making the decision to
capitalize on the opportunities and
mitigate the threats. This will be
explained through user-interaction
and the decisions of human operators, in order to ultimately facilitate
proactive decision making.

Applications
The proposed methodology for proactive event-driven decision making
is applicable to a wide range of application domains. For example, traffic
management and credit card fraud
management are two applications.
To illustrate the methodology,
proactive traffic management can be
attained by following the steps below.

IEEE TECHNOLOGY AND SOCIETY MAGAZINE

FALL 2014

Table of Contents for the Digital Edition of IEEE Technology and Society Magazine - Fall 2014