Jaekoo Joo

Agent-based simulation of affordance-based human behaviors in emergency evacuation

Abstract Complex cognitive processes corresponding to human control behaviors cannot be easily inferred using (1) a logical rule-based model, (2) a statistical model, or (3) an analytical predictive model. Predicting human behaviors in complex and uncertain environments like emergency evacuation is considered almost impossible (at least NP hard) in systems theory. In this paper, we explore simulating human behaviors using affordance-based finite state automata (FSA) modeling, based on the ecological concept of affordance theory. To this end, we introduce the conceptual and generic framework of affordance-based human behavior simulation developed through our previous work. Following the generic framework, formal simulation models of affordance-based human behaviors are developed, especially for emergency evacuation, to mimic perception-based dynamic human actions interacting with emergent environmental changes, such as fire. A “warehouse fire evacuation” case is used to demonstrate the applicability of the proposed framework. The human action planning algorithms in the simulation model are developed and implemented using the Adjusted Floor Field Indicators, which represent not only the evacuee’s prior knowledge of the floor layout but the perceivable information about dynamic environmental changes. The results of our simulation study verify that the proposed framework accurately simulates human fire evacuation behavior. The proposed framework is expected to capture the natural manner in which humans behave in emergency evacuation and enhance the simulation fidelity of analyses and predictions of perceptual human behaviors/responses in the systems by incorporating cognitive intent into human behavior simulations.

Adaptive and dynamic process planning using neural networks

Although feature-based process planning plays a vital role in automating and integrating design and manufacturing for efficient production, its off-line properties prevent the shop floor controller from rapidly coping with dynamic shop floor status such as unexpected production errors and rush orders. This paper proposes a conceptual framework of the adaptive and dynamic process planning system that can rapidly and dynamically generate the needed process plans based on shop floor status. In particular, the generic schemes for constructing dynamic planning models are suggested. The dynamic planning models are constructed as neural network forms, and then embedded into each process feature in the process plan. The shop floor controller will execute them to determine machine, cutting tools, cutting parameters, tool paths and NC codes just before the associated process feature is machined. The dynamic nature of process planning enables the shop floor controller to increase flexibility and efficiency in unexpected situations.

Efficient sculptured pocket machining using feature extraction and conversion

Abstract A methodology is presented for finding a feature that can be used from design to manufacturing for sculptured pockets. A feature is the core concept necessary to realize a fully integrated CAD/CAM system; the information contents embedded in the feature can be easily conveyed from one application to another. However, the feature generated in one application may not be directly suitable for another without being modified with more information. This paper presents a methodology for decomposing a bulky feature of a sculptured pocket into compact features to be efficiently machined. In particular, the paper focuses on two tasks: (1) to horizontally segment a bulky feature into intermediate features and to generate their temporal precedence graph, and (2) to further decompose each intermediate feature vertically into smaller manufacturing features and to apply the variable feed rate to each small feature. The proposed method will provide better efficiency in machining time and cost than the classical method, which uses a long string of NC code.

An affordance-based formalism for modeling human-involvement in complex systems for prospective control

We propose a predictive modeling framework for human-involved complex systems in which humans play controlling roles. Affordance theory provides definitions of human actions and their associated properties, and the affordance-based Finite State Automata (FSA) model is capable of mapping the nondeterministic human actions into computable components in modeling formalism. In this paper, we further investigate the role of perception in human actions and examine the representation of perceptual elements in affordance-based modeling formalism. We also propose necessary and sufficient conditions for mapping perception-based human actions into systems theory to develop a predictive modeling formalism in the context of prospective control. A driving example is used to show how to build a formal model of human-involved complex system for prospective control. The suggested modeling frameworks will increase the soundness and completeness of a modeling formalism as well as can be used as guide to model human activities in a complex system.

Dynamic planning model for determining cutting parameters using neural networks in feature-based process planning

Although feature-based computer-aided process planning plays a vital role in automating and integrating design and manufacturing for efficient production, its off-line properties prohibit the shop floor controllers from rapidly coping with unexpected production errors. The objective of the paper is to suggest a neural network-based dynamic planning model, by which the shop floor controllers determine cutting parameters in real-time based on shop floor status. At off-line is the dynamic planning model constructed as a neural network form, and then embedded into each removal feature. The dynamic planning model will be executed by the shop floor controllers to determine the cutting parameters. A prototype system is constructed to validate whether the dynamic planning model is capable of determining dynamically and efficiently the cutting parameters for a particular set of shop operating factors. Owing to the dynamic planning model, the shop floor controller will increase flexibility and robustness by rapidly and adaptively determining the cutting parameters in unexpected errors occurring.

A modelling formalism for human-machine cooperative systems

This paper presents a collection of models of humans involved in complex systems with a focus on control of the system while allowing human participation in decisions for system operation. The existence of a formal model that captures human behaviour in a complex system allows for the efficient development of modelling and control software of man-machine systems. This paper provides a foundation for modelling and software development for complex systems that includes human activities. A modelling formalism, based on the automata theory for human-machine cooperative systems is demonstrated in this paper in consideration with the ecological concept of affordance.

Neural Network-based Dynamic Planning Model for Process Parameter Determination

Determining such process parameters as rotational speed, feed rate, depth of cut, and width of cut is the critical function that affects not only machining productivity but also quality of a finished part. In the paper, a dynamic planning model is developed to determine efficient process parameters for roughly machining a pocket type of process feature in the shop floor. A neural network structure is proposed to implement the dynamic planning model. The learning patterns used for training the neural network are acquired through simulation-based optimization procedures. The procedure finds an optimal set of process parameters for each set of operating factors by considering the machining costs, cutting forces, and machining power. A prototype system is developed and experimented to demonstrate the feasibility of the proposed model. Due to the dynamic planning model approach, the fatal weaknesses of conventional processing parameter determination can be conquered by its efficient, dynamic, and adaptive planning ability

Perception and BDI Reasoning Based Agent Model for Human Behavior Simulation in Complex System

Modeling of human behaviors in systems engineering has been regarded as an extremely complex problem due to the ambiguity and difficulty of representing human decision processes. Unlike modeling of traditional physical systems, from which active humans are assumed to be excluded, HECS has some peculiar characteristics which can be summarized as follows: 1) Environments and human itself are nondeterministic and dynamic that there are many different ways in which they dynamically evolve. 2) Human perceives a set of perceptual information taken locally from surrounding environments and other humans in the environment, which will guide human actions toward his or her goal achievement. In order to overcome the challenges due to the above characteristics, we present an human agent model for mimicking perception-based rational human behaviors in complex systems by combining the ecological concepts of affordance- and the Belief-Desire-Intention (BDI) theory. Illustrative models of fire evacuation simulation are developed to show how the proposed framework can be applied. The proposed agent model is expected to realize their potential and enhance the simulation fidelity in analyzing and predicting human behaviors in HECS.

Modeling and Simulation of Emergent Evacuation Using Affordance-based FSA Models

Modeling and simulation of human-involved complex systems pose challenges to representing human decision makings into logical systems because of the nondeterministic and dynamic nature of human behaviors. In modeling perspectives, human`s activities in systems can increase uncertainty and complexity, because he or she can potentially access all other resources within the system and change the system states. To address all of these human involvements in the system, this research suggests applying the Finite State Automata (FSA)-based formal modeling of human-involved systems that incorporates the ecological concept of affordances to an evacuation simulation, so that human behavioral patterns under urgent and dynamic emergency situations can be considered in the real-time simulation. The proposed simulation methodologies were interpreted using the warehouse fire evacuation simulation to clarify the applicability of the proposed methodology. This research is expected to merge system engineering technologies and human factors, and come out to the new predictive modeling methodology for disaster simulations. This research can be applied to a variety of applications such as building layout designs and building access control systems for emergency situations.

Ontology for Supplier Discovery in Manufacturing Domain

Discovering the suppliers capable of manufacturing the parts that satisfy buyer requirements via current online market places remains difficult due to semantic differences between what the suppliers can produce and what the buyer wants to acquire. One of the promising approaches to overcome the semantic diffe-rence is to adopt an ontology to describe the suppliers’ manufacturing capabilities and the buyer requir-ements that range widely from manufacturing costs to eco-friendly design. Such an ontology dedicated to supplier discovery has yet to be developed. MSDL(Manufacturing Service Description Language) provides the basis for defining terms and their relationships in the ontology. Thus, the objective of this paper is to extend MSDL into a new ontology suitable for supplier discovery in mold manufacturing industry. In addition, a new ontology development method for supplier discovery will be proposed. Finally prototype demonstrations are provided to show a feasibility of the proposed ontology in mold manufacturing domains.