Carlos Herrera

Viable System Model Approach for Holonic Product Driven Manufacturing Systems

The chapter presents a generic framework for a Product Driven Control Systems (PDCS) dealing with production planning and control. The framework is based on Viable System Model (VSM) which is introduced for intelligent manufacturing systems. Based on suitable properties as autonomy, self-organization and adaptability, VSM allows modelling and considering these properties for PDCS. An application for a Manufacturing Planning and Control System (MPCS) is proposed. Different points of view are also presented based on this application which are shown and explained in the sense of VSM principles. Finally, a discussion is presented dealing with the main issues of the proposed approach.

A reactive decision-making approach to reduce instability in a master production schedule

One of the primary factors that impact the master production scheduling performance is demand fluctuation, which leads to frequently updated decisions, thereby causing instability. Consequently, global cost deteriorates, and productivity decreases. A reactive approach based on parametric mixed-integer programming (MIP) is proposed that aims to provide a set of plans such that a compromise between production cost and production stability is ensured. Several stability measures and their corresponding MIP model are proposed. An experimental study is performed to highlight the effectiveness of the reactive approach with regard to the proposed performance measures. It is observed that an improvement in stability does not mean a significant increase in the total production cost. Furthermore, the procedure yields a set of plans that in practice would enable flexible management of production.

Graphene oxide: An efficient material and recent approach for biotechnological and biomedical applications

The two-dimensional (2D) derivative of graphite termed graphene has widespread applications in various frontiers areas of nanoscience and nanotechnologies. Graphene in its oxidized form named as graphene oxide (GO) has a mixed structure equipped with various oxygen containing functional groups (epoxy, hydroxyl, carboxylic and carbonyl etc.) provides attachment sites to various biological molecules including protein, deoxyribonucleic acid (DNA), ribonucleic acid (RNA) etc. The attached biological molecules with the help of functional groups make it a promising candidate in research field of biotechnological and biomedical applications. The ease of processability and functionalization in aqueous solution due to available functional groups, amphiphilicity, better surface enhanced Raman scattering (SERS), fluorescence and its quenching ability better than graphene make GO a promising candidate for various biological applications. The amphipathetic nature and high surface area of the GO not only prepare it as a biocompatible, soft and flexible intra/inter cellular carrier but also provides long-term biocompatibility with very low cytotoxicity. Inspite of this, still we lack a very recent review for advanced biological applications of graphene oxide. This review deals the bio application of GO and the recent advancement as a biosensors, antibacterial agent, early detection of cancer, cancer cell imaging/mapping, targeted drug delivery and gene therapy etc.

Automatic generation of algorithms for the binary knapsack problem

Because it is classified as NP-hard, the binary knapsack problem is a good example of a combinatorial optimization problem that still presents increased difficulty when attempting to determine the optimal solution for any instance. Although exact and heuristic methods have been developed in an attempt to solve the problem, such methods have been unable to solve even small instances of the problem. In this paper, new algorithms for this problem are automatically generated by means of genetic programming from sets of training instances of different sizes and are then evaluated against other larger sized sets of instances, thereby detecting the robustness of the algorithms for larger instances. Overall, the produced algorithms are able to identify up to 52% of the optimal solutions for the biggest instances used.

A product-driven system approach for multilevel decisions in manufacturing planning and control

Decision-making in Manufacturing Planning and Control Systems uses processes that consider several levels of product aggregation and different time horizons. Decisions rendered on each level do not always have similar goals. The problem is that building Intelligent Manufacturing Systems (IMS) involves coordinate decisions on different levels to achieve a common objective. There is no current research on IMS regarding coordination among decision levels in Product-driven Control Systems, hence simulations of the planning and control processes should be conducted to analyze the behavior of multilevel objectives. Accordingly, in this paper a simulation model gives account of different coordination issues between tactical and operational levels. At the tactical level, production plans are generated by an Advanced Planning and Scheduling system, whereas at the operational level, a distributed decision rule is used. The whole simulation makes decentralized decisions that are managed by production lots represented by holons. Results for our simulation indicate that coordination among active lots is capable of making effective multilevel distributed decisions when compared with conventional approaches.

A Holonic Manufacturing System for a Copper Smelting Process

In this article, a holonic manufacturing system to coordinate the activities in a copper concentrate smelting process is designed. The system was developed and compared with a simulation of the current system in use. Four scenarios with different feeding levels of copper concentrate were considered. The results show an increase in the amount of processed ore in the four studied scenarios because of a reduction in the total waiting time and more efficient use of the plant equipment. Thus, the better use of resources and a decrease in idle time because of improved coordination of activities and different components of the system is highlighted.

A Case-Based Reasoning Approach to Find Good Compiler Optimization Sequences

Among several optimizations provided by an optimizing compiler, it is a challenge, even for the most expert programmer, to know which compiler optimizations will generate the best target code. The goal of this paper is to describe a case-based reasoning approach that automatically selects a compiler optimization sequence that is able to outperform a well-engineered compiler optimization level in terms of runtime. The results obtained by our approach indicate that it achieves improvement on a set of standard benchmarks over a well-engineered compiler optimization sequence.

Simulation of a Hybrid Product-Driven System for Manufacturing Planning and Control

Manufacturing planning and control systems (MPCS) incorporate processes that consider several levels of product aggregation and different time horizons for decision making. The decisions rendered on each level do not always have similar objectives. In the context of intelligent manufacturing systems (IMS), the coordination of decisions on different levels is a fundamental problem. Extensive research on IMS, specifically regarding coordination among decision levels in product-driven control systems (PDCS), is nonexistent. Therefore, simulations of the planning and control processes are proposed to analyze the coordination of multilevel objectives. The proposed implementation simulates the coordination between tactical and operational levels. At the tactical level, production plans are obtained through a system based on advanced planning and scheduling (APS). At the operational level, a decentralized system, which is based on distributed decision rules, is implemented. The simulation considers decentralized decisions that are managed by production lots, which are modeled as holons, and based on an industrial study case. The results indicate that coordination is feasible and highlight the importance of the reactivity caused by the distributed decisions made by the active lots. The proposed simulation schema can also be used to compare conventional and holonic collaborative approaches.