Pierpaolo Caricato

Parallel tabu search for a pickup and delivery problem under track contention

This article introduces the Pickup and Delivery Problem under Track Contention, a particular vehicle routing problem in which loads have to be transported between origin-destination pairs by means of vehicles travelling along a capacitated network. Two sequential heuristics and a parallel tabu search are proposed. Computational experiments show that the parallel tabu search is able to find much better solutions than the sequential procedures, although this comes at the expense of a higher computing time.

An online approach to dynamic rescheduling for production planning applications

Manufacturing firms must often consider how to plan the production of a new order while containing the impact on the existing plan. Two approaches are typically used to solve the problem: online scheduling and rescheduling. State-of-the-art for both strategies is analyzed and methods available in the literature are proved to be inefficient for real world cases coming from production planning problems in a manufacturing firm. We propose an alternative approach combining the most effective aspects of both traditional approaches. The already available production plan and the characteristics of the new order to be planned are studied and used in order to generate a new production plan that meets two requirements: containing the number of changes to the existing plan and minimizing the delays due to the newly planned order. Constraint programming is used to implement the proposed approach. Results on case studies are also provided to evaluate the effectiveness of the proposed approach.

Using simulated annealing to design a material-handling system

Among all material-handling systems used in modern factories, those based on automated guided vehicles have, for many reasons, met with great success. AGV systems are much more flexible than traditional systems such as fixed-path conveyors, allowing dynamic reconfiguration of guidepaths in accordance with factorys changing transportation needs.

Improved formulation, branch-and-cut and tabu search heuristic for single loop material flow system design

The single loop material flow system design is a combinatorial optimization problem, arising in material handling system design, which amounts to designing an unidirectional loop flow pattern as well as to locate pickup and delivery stations. The objective is to minimize the time required to carry out all material flow movements between cells. In this paper, we develop valid inequalities for a previously proposed formulation. The valid inequalities are then embedded into a branch-and-cut framework which is shown to solve much larger instances to optimality than those reported in the literature. A tailored tabu search heuristic is also illustrated and computationally assessed.

Simulation and mathematical programming for a multi-objective configuration problem in a hybrid flow shop

This paper introduces an application of simulation-based multi-objective optimization to solve a system configuration problem in a hybrid flow shop system. The test case is provided by a firm that manufactures mechanical parts for the automotive sector. We present an architecture that uses both discrete-event simulation and mathematical programming tools in order to solve the problem. The multiple-objective nature of the problem is preserved throughout the proposed approach, using Pareto-dominance concepts both to eliminate inefficient solutions within the proposed solution algorithm and to provide the user with efficient solutions. Mathematical programming is used to cull the required number of simulation runs. Computational results obtained using a real-world case study are reported. The proposed approach is benchmarked against a general purpose simulation-optimization engine in order to prove its effectiveness.

Long-term planning in manufacturing production systems under uncertain conditions

Nowadays, the frequency of decisions related to the configuration and capacity evaluation of manufacturing production systems is increasing in more and more industrial sectors, especially in the automotive field. This is due to a variety of factors, such as the reduction of the life cycle of the product, increasing competition, etc. In such a context, decision makers have to take their actions in shorter times than they ever did in the past: as an example, they typically need to take quick decisions about different production system alternatives. This specific problem has increased in complexity because of the necessity to take into account all the sources of variability and each related level of uncertainty in the available data definition. Two main aspects lead to such difficulties: the lack of a proper decision support system and the need to contextually model the uncertain data. This paper presents the first step in this direction. In particular, a decision support system (DSS) has been developed to help decision makers take productive capacity planning decisions according to the uncertain characterisation of the market evolution. First, a strategy evaluation tool allows the decision maker to specify several productive capacity expansion policies and, then, uses a fuzzy discrete event simulation paradigm (Fuzzy-DEVS) to compare them, providing the possibility of choosing between the different alternatives according to performance indicators. A strategy design tool helps the decision maker by inferring the best expansion policy on the basis of the system analysis conducted in the first step. Finally, our approach has been validated by means of an industrial test case in the automotive sector.

Birth and Evolution of a Decision Support System in the Textile Manufacturing Field

We present the evolution of a Decision Support System that was developed for a company that produces high-tech fabrics. The project started in 2010 within a cooperation agreement between a public Italian university and the firm, initially addressing what was perceived as the main and more peculiar aspect of the decision process related with the production planning: namely the sheet booking process. We designed and developed a DSS that implemented a mathematical programming algorithm based on combinatorial optimization to solve the very peculiar variant of the cutting stock problem that could be used to model the decision process. The results of the usage of the DSS outperformed any estimation the company had expected from the project, leading not only to the automation of the decision process, but also to a large enhancement in the material usage rates. We present a short outline of the improvements achieved with this first tool. The positive results obtained with the first DSS led to two further evolutions of the tool: the former was developed 2 years later, while the latter is currently being developed.

The Integration of Decision Analysis Techniques in High-Throughput Clinical Analyzers

From the early 1990s, the introduction of high-throughput clinical analyzers has significantly changed the workflow of In-Vitro-Diagnostics (IVD) tests. These high-tech instruments have helped and keep helping clinical laboratories both to increase quality diagnostic responses and to get more for every dollar they spend. Nevertheless, IVD industrial research has been up to now largely hardware-driven with the introduction in the market of many sophisticated technologies. The software component, models and decision support systems in particular, has lagged behind. To reach the full potential of diagnostic automation, it must be addressed the challenge of making the most intelligent use of the hardware that is deployed. Focusing on time efficiency, the authors have devised an operations research-based method for a class of high-throughput clinical analyzers. To demonstrate the validity of the research, the proposed method has been coded and integrated into the Laboratory Information System of the Laboratorio di Analisi Cliniche Dr. P. Pignatelli, one of the most important clinical laboratories in Southern Italy. Siemens Immulite®; 2000 has been the reference case. The enhanced operating planning procedure provides a monetary benefit of 52,000 USD/year per instruments and a trade-off between clinical benefits and operating costs equivalent to the one provided by the current hardware-driven research at Siemens. Despite the proposed approach has the potential to determine guidelines for enhancing a wide range of current high-throughput clinical analyzers, we have to register a failure in trying to convince technology providers to invest in embedding such new models in their hardware. Some possible causes for such failure are highlighted, trying to find possible improvements for future developments.

Augmented Reality Applications in Manufacturing: A Multi-Criteria Decision Model for Performance Analysis

Abstract Augmented Reality (AR) applications are becoming mature technologies for the use in manufacturing systems. Their very innovative character together with the variety of devices are now forcing production managers and researchers to analyse their application from technological to organizational point of view. The aim of the paper is to propose a multi-criteria model which integrates technical and organizational metrics to provide reliable decision support system for analysing the application of AR technologies in manufacturing. The proposed model applies the AHP (Analytic Hierarchy Process) method for integrating effectively technological and organizational factors which will contribute to analyse how an AR system could be effectively applied in the manufacturing sector.

Selecting Capacity Plan

Nowadays, the frequency of decisions related with the configuration and capacity evaluation of manufacturing production system is increasing in more and more industrial sectors. In such a context, decision-makers have to take their actions in shorter times than they ever did in the past. This problem has increased in complexity because of the necessity to take into account all the sources of variability and each related level of uncertainty in the available data definition. For such a reason the capacity plan selection process is still an open question. In this chapter, a Decision Support System has been developed to help decision-makers to take productive capacity planning decisions according to uncertain characterization of the market evolution. The proposed methodology can be used to take strategic decisions over a long term programming horizon, allowing an effective comparison of user-defined strategies according to user-defined efficiency parameters. In the proposed approach, the strategy expansion evaluation concerns the designing of the state evolution, the representation of the system dynamic and the research of the suitable capacity plan. Finally, our approach has been validated by means of the reference case study.