Intermodal Mobility Assistance for Megacities - Smart Information Systems: Computational Intelligence for Real-Life Applications - page 342

find appropriate routes through the city, with a focus on avoiding congestion and
traffic jams.
The aim of this chapter is twofold. First, we present a mechanism that allows
combining different and heterogeneous information channels. For this purpose, we
make use of the agent paradigm [ 26 ] as well as of multiagent system architec-
tures [ 25 ].
Second, we describe selected information channels in more detail. Our intention is
to outline a system architecture that can be used to integrate, preprocess, and utilize
heterogeneous traffic information from different channels and to show how these
channels have to be implemented in order to make the approach work.
13.2 The IMA System—An Overview
The IMA system provides solutions to its users for effective transportation. The aim
of the system is to increase the quality of life in megacities. While IMA is a complex
software application, which comprises dozens of services, components, apps, front-
ends, etc., the focus of this chapter is on IMA's routing capability ( IMA routing ). IMA
routing is principally implemented through two fundamental components, namely:
(1) the Route Planning component and (2) the Video Analysis component. In this
section, we elaborate on the architecture of our system. In order to do so, it is nec-
essary to provide details about the framework that was used for its implementation,
which can be found in Sect. 13.2.1 . Subsequently, in Sect. 13.2.2 , we describe the
architecture of our system.
13.2.1 IMA—A Distributed System
The IMA system aims to integrate different services, which are not necessarily
running on the same hardware. Thus, from an engineering perspective, the IMA appli-
cation is a distributed system, which can be considered as multiagent system [ 25 ].
Agent-oriented software engineering [ 12 ] is a common approach to develop such
distributed systems. The advantage of applying an agent-oriented view for the devel-
opment is the presence of a comprehensive set of methodologies, mechanisms, and
tools, which significantly eases the engineering process.
One particular tool, which facilitates the development of agent-based systems is
the Java-Based Intelligent Agent Componentware ,or JIAC [ 16 ]. The development
of JIAC was geared toward reliability and robustness [ 17 ]. As both characteristics
are vital for our system (which is meant to be deployed in a real-life scenario), JIAC
appeared to be a judicious choice as the basis of our implementation.
The JIAC architecture is a concept that builds on multiple agents, with each agent
serving as an intelligent service execution container. Thus, a multiagent system can
work as a flexible and dynamic service execution platform that corresponds to the