An agent-oriented approach to resolve the production planning complexities for a modern steel manufacturing system

Abstract

The development of strategies for sequencing slabs through the reheat furnace and rolling mill of a modern steel manufacturing system is complicated by multiple and often conflicting objectives. For example, optimal energy efficiency through the reheat furnace may lead to inefficient rolling sequences or a less than desirable product delivery schedule. Thus, not only the model formulation is complicated, but also the combinatorial nature of the problem precludes an optimal solution. To address these complexities, an agentoriented approach that has originated from distributed artificial intelligence (DAI) is proposed here. It is found that the implementation of agent technology in the steel manufacturing system makes the operations more flexible, economical and energy efficient.

Identifying Product and Process State Drivers in Manufacturing Systems Using Supervised Machine Learning

“…stands for in general and having introduced the main techniques, next, a more detailed discussion will focus on implications of a manufacturing process and its relation to quality. When looking at improving manufacturing processes, as a first step towards efficient manufacturing, it has to be ensured that the manufacturing processes, the entire manufacturing programme for that matter, design is capable to produce the desired product properties (Mohanty 2004). After this overarching requirement, a functional process design, is given, the process quality plays a major role, as it is directly connected to product quality Jacob and Petrick 2007).…”

“…To do so, it has to be confirmed that these desired product properties can be achieved through the production process (Mohanty 2004). Therefore, a detailed understanding of the manufacturing programme, its processes/operations and their influence on products and in the end, their product state, becomes progressively more important.…”

Integrated Scheduling of Reheating Furnace and Hot Rolling Based on Improved Multiobjective Differential Evolution

“…However, the mathematical model for the problem was not provided. Mohanty [3] developed an agent-based heuristic to determine the slab sequence in the reheating furnace and hot rolling; however, the overheating and sequence-dependent cost were not considered. Wang and Tang [4] formulated the scheduling of a single continuous walking beam reheating furnace as a mixed-integer linear programming model and presented an improved particle swarm optimization (PSO) algorithm to obtain a near-optimal solution.…”

Development of a collaborative virtual maintenance environment with agent technology

“…For multi-agent technology as the other research stream, even if there is much literature in various areas, we concentrate on the agent-based application in mechanism engineering fields, such as in product design process [23][24][25], in process planning process [26][27][28], in fixture design process [29], in production planning and scheduling process [30][31][32][33], in product maintenance process [34], in supply chain system [35], in workflow system [36,37] etc. A few researchers apply agent technology to manufacturing system [38][39][40], such as manufacturing resource management [41], manufacturing task assignment [42], manufacturing system control [43], manufacturing knowledge management [44], manufacturing system integration [45] etc.…”