Building a rigorous foundation for performance assurance assessment techniques for “smart” manufacturing systems

Abstract

The highly networked and real-time data analysis features of smart manufacturing systems (SMS) require different information infrastructure, data analytics technology, and performance assurance methodologies. The main purpose of this paper is to (i) explore the complete product-process performance assurance space to identify the key performance indicators that help evaluate and quantify system performance at different abstraction levels, (ii) discuss models and methodologies for data analytics, and (iii) suggest a digital factory-based simulation technique to evaluate those key indicators for performance prediction. The paper presents a systematic and rigorous approach towards establishing these performance assurance methodologies applicable to complex value chains of smart manufacturing systems by extensively exploring all possible product and process related performance issues. A hypercube information model is proposed for the purpose of formal representation of the highly dimensional and correlated information among different actors in a smart manufacturing system, thus providing a rigorous foundation for the performance assurance space. The relevant taxonomy and an ontology-based framework are then developed for formal representation of the entities, activities and knowledge involved in the performance assurance domain. It provides a detailed insight into the PA space and defines appropriate measures which can be applied to predict and improve system performance assurances.There remains an urgent need for developing general methodology and standardized metrics for the wider use of the metrics by different manufacturers in different manufacturing domains. There have been many attempts in building such methodologies. Feng and Joung [4] presented a comprehensive list of previous approaches. Attempts were also made in the area of the complex production system to

A functional modeling approach for quality assurance in metal additive manufacturing

“…The objectives and constraints help determine the materials, resources, machines, software, operator and budget cost. Meanwhile, the KPIs help determine various performance measures, including asset utilization, agility and sustainability (Roy et al , 2014). Asset utilization is related to planning and maintenance activities.…”

A taxonomy for key performance indicators management

“…And not only that, as we will see later, there has been few effort towards the definition of a structured knowledge organization of KPIs. As stated in [8], the absence of a common terminology and an organized structure of 40 knowledge items can be an obstacle to the communication and understanding of the produced research-related information, making works harder to analyze and compare, or even find interrelationships among such works. This issue can be an impediment to both the progress in research and the transfer of research results to the market.…”

“…As a consequence, each KPI is classified following two criteria, one related with the above-mentioned domain (for example, quality, time or cost), and another 285 related with the production and manufacturing process (design, manufacturing, environmental or customer). An approach with certain similarities to the previous one is that presented in [40], in which authors propose a taxonomy of KPIs where two levels are considered. In this case, besides aspects related with the domain, KPIs are classified into six categories: manufacturing, logistics, 290 personnel, financial, supply chain, or learning and innovation.…”

“…Dependencies between KPIs can be explicitly specified representing, for example, the components used in the computation formula [36]. This kind of relationships among components can lead to basic KPIs, compound or derived [2,3,6,19,24,31,36,37,39,40,45]. As other examples, the self-explanatory pair "givesDataTo" and "takesDataFrom" relationships are defined in [30].…”

“…[19,2], together with the unit of measure (percentage, units of weight, length, currency, etc.) [36,40]. The ISO 22400 standard defines the Unit of measure of each KPI as the "basic unit or dimension in which the KPI 365 is expressed".…”