978-1-4799-5937-2/14/$31.00 ©2014 IEEE 145
Abstract - Current manufacturing scheduling has still
difficulties to deal with real-world situations and, hence, human
intervention is required to maintain real-time adaptation and
optimization, to efficiently adapt to the inherent complex system
dynamics. In this paper the prototype of an Adaptive Decision
Support System for Interactive Scheduling with MetaCognition
and User Modeling Experience (ADSyS) is proposed. A
preliminary usability evaluation was streamlined to collect
user’s opinion about the system performance and interaction
model.
Keywords – Dynamic Scheduling; Human-Computer Interaction;
User Modeling; Learning; Optimization; Usability.
I. INTRODUCTION
challenge is emerging in the design of scheduling
decision support systems for manufacturing
environments, where dynamic adaptation and
optimization is becoming increasingly important in
incorporating expert’s knowledge and experience.
Scheduling is one of the most interesting and complex
problems, leading to a significant effort from the R&D
community. Several optimization tools have been developed
and some of them were applied to real world Manufacturing
Systems. Decision Support Systems (DSS) and Expert
Systems are two approaches widely used to generate
schedules using experience or expert knowledge. Current
manufacturing scheduling still has difficulties in real-world
situations and, hence, human intervention is required to
maintain real-time adaptation and optimization, to efficiently
adapt to the inherent complex system dynamics.
This work investigates the development of
Learning-Assisted Intelligent Scheduling Systems that use
active learning by experience or even by observation of
expert's decisions. It will be provided with the ability to
define its intelligence and the user is able to select the level of
experience according to business requirements. The
motivation for this arises from the requirement of developing
approaches that should be able to adaptively control,
coordinate and optimize, addressing the several challenges
presented on complex real-world problems (RWP). By using
advanced algorithms, modeling and services, we intend to
provide the decision maker with the conditions to make the
right choices when and where they are significant for the
business.
Regardless the numerous advances in DSS and Multi-agent
systems (MAS) for scheduling [1-14], most development
systems still require that users have extensive knowledge of
programming techniques, as well as the specific computer
system at hand. At this point, we identify as relevant for this
project the integration of the paradigms: Human-Computer
Interaction (HCI), DSS for scheduling, MAS, Machine
Learning (ML), Autonomic Computing (AC) and Nature
Inspired Optimization (NIO).
The prototype of an Adaptive Decision Support System for
Interactive Scheduling with MetaCognition and User
Modeling Experience (ADSyS) is proposed. The system
architecture is explained and the preliminary usability
evaluation considering interaction is described.
The remaining sections are organized as follows: section II
summarizes some literature review on Real World
Scheduling, Human-Computer Interaction and User
Modeling. Section III describes the architecture for Adaptive
Decision Support System for Interactive Scheduling with
MetaCognition and User Modeling Experience. In section IV,
the prototype usability evaluation is exposed. Finally, the
paper presents some conclusions and puts forward some ideas
for future work.
II. LITERATURE REVIEW
This section presents some insights about Real-World
Scheduling, Human-Computer Interaction and User
Modeling.
A. Real-World Scheduling
Scheduling function in a manufacturing organization is
outlined and integrated in the global management decision
system, considering the diverse organizational levels, their
functional perspective or the integration of functions and
business processes. DSS are considered essential to
adequately support globalization strategies and reengineering
business processes toward the achievement of competitive
advantages, with impact on cost reduction, differentiation and
innovation strategies, promoting and facilitating relations
with business partners and potential customers.
Scheduling, as a decision-making process, plays an
important role in most manufacturing systems as well as in
most information processing environments.
“MANUFUTURE-a vision for 2020” is a strategic
document summarizing the High Level Group’s (European
Commission) recommendations, identifies some concerns at
ensuring the long-term competitiveness of the EU
Prototype of an Adaptive Decision Support System for Interactive
Scheduling with MetaCognition and User Modeling Experience
A. Madureira, S. Gomes, B. Cunha, J.P. Pereira, J. M. Santos, I. Pereira
GECAD - Knowledge Engineering and Decision Support Research Center
School of Engineering from Polytechnic of Porto (ISEP/IPP)
Porto, Portugal
{amd,srpgo,bmaca,jjp,jms,iaspe}@isep.ipp.pt
A