Manufacturing Execution Systems for lean, adaptive production processes

In order to deal with global competition, industries have undertaken many efforts directed to improve manufacturing efficiency. From a broad perspective, the adopted approaches could be classified in two categories: 1. the simplification of manufacturing processes and relative control systems, leading to lean manufacturing methododologies and techniques; 2. the massive deployment of information tools and computational algorithms, aiming to plan and control all the activities in detail, in spite of system complexity. For several years, these two approaches have been assumed to be mutually exclusive; nevertheless, information collection and analysis are mandatory to define improvement strategies and assess their impact; therefore, the deployment of lean manufacturing methodologies cannot exclude the integration of Information Technology (IT) tools. The aim of this work is to investigate on methodologies and techniques adoptable to improve the efficacy of Manufacturing Execution Systems (MES), a class of software that allows data exchange between the shop-floor and the organizational levels, enabling the implementation of the lean manufacturing approach. Today, the feedback information in the available MES mainly consists in key performance indicators, such as cycle time, work in process and resources utilization. Beside this, MES requires the integration of functionalities for process monitoring and control, aiming at the reduction of wastes and supporting continuous improvement. Hence, mathematical techniques able to analyze data in real-time and provide useful information to adaptively control the process are studied in this work. To provide the evidence of the feasibility and effectiveness of the approach, as well as the independence from any specific manufacturing technology, different case studies, both in the fields of subtractive and additive manufacturing, have been developed. In the former, a technique for the automatic alignment of a spur gear has been studied: geometrical measurements are acquired and analyzed in real-time to provide the values for two feasible part rotations resulting in the gear configuration with minimum positioning error. Such gears are manufactured for applications in aeronautics, and the deployment of this automation system is particularly significant because of the tight tolerances to be satisfied. The latter case study deals with a Fused Deposition Modeling process: an algorithm able to monitor part surface accuracy and identify defects has been developed. This methodology allows to evaluate in real-time whether the quality of the part is satisfactory or not; in case of negative response, the process can be stopped avoiding material loss. The implemented techniques enable product quality improvement, as well as the reduction of wasted material and time. Nevertheless, the deployment of such information only for process control purposes is restrictive; a framework to use this knowledge for supporting the design and the continuous improvement of a product or a process is presented. Furthermore, two case studies have been dealt to extend the application of MES tools from manufacturing operations to ancillary services. The first one is in the field of automated warehouses: a combined approach made of mathematical models and simulations has been developed. Analytical tools have been defined to evaluate the average performance of a system in simple, pre-determined situations; conversely, the simulation tool aims at a higher detail level of assessment, since in the real shop-floor deployment, different, composite storage and retrieval activities can take place. In the second case-study, mathematical models and simulation are used to support the re-design of a manufacturing process; the focus is on the transport of items through the line, performed by automated vehicles. The mathematical model has been developed to identify the optimal layout of the workstations; simulations are used to evaluate the tasks to be performed by the automated vehicles and the resulting performance. In both the applications, the deployment of simulation tools allows to evaluate complex or even unexpected scenarios by predicting the behavior of a system, preventing criticalities, and evaluating the impact of a change in the process. The management criteria can be adapted according to the features of the real situation to be faced; this leads to better exploit the available resources, to improve productivity and identify waste sources, consistently with the lean paradigm.