artificial intelligence blockchain building energy management building information modelling data analytics data mining energy efficiency Energy efficient building energy management energy performance indicator flexibility measurement industry 4.0 industry 4.0 maturity assessment Internet of Things knowledge management linked data machine learning manufacturing Ontology ontology engineering ontology population product configuration product lifecycle management production planning and control requirement engineering resource efficient manufacturing smart cities Supply Chain 4.0 sustainability virtual engineering
2020 |
Wicaksono Hendro; Ni, Tianran An Automated Information System for Medium to Short-Term Manpower Capacity Planning in Make-To-Order Manufacturing Journal Article Procedia Manufacturing, 52 , pp. 319-324, 2020. Abstract | Links | BibTeX | Tags: information systems, Information visualisation, production planning and control, resource efficient manufacturing @article{Wicaksono2020d, title = {An Automated Information System for Medium to Short-Term Manpower Capacity Planning in Make-To-Order Manufacturing}, author = {Wicaksono, Hendro; Ni, Tianran}, url = {https://doi.org/10.1016/j.promfg.2020.11.053}, doi = {10.1016/j.promfg.2020.11.053}, year = {2020}, date = {2020-12-31}, journal = {Procedia Manufacturing}, volume = {52}, pages = {319-324}, abstract = {In today’s tough economy, it is important for (Make-To-Order) MTO companies to be responsive to customer demand and market fluctuations and to keep the costs as low as possible at the same time. Unlike in Make-To-Stock (MTS), MTO companies hold capacity in reserve. Thus, they are able to make efficient utilization of available capacity to satisfy customer needs. This then leads to a constant capacity planning problem. The companies are facing fluctuations between overload by lack of sufficient capacity, and idleness by excess of capacity comparing to the level of demand. Among all the planning resources, the available manpower is one of the most essential parts of the MTO operations. Therefore, the allocation and adjustment of manpower capacity that suits different planning horizons is a predominant measure to meet the changing capacity demands. Nonetheless, a signing each individual labor to various types of tasks and orders on a day to day basis continually for the planning horizon of several weeks or months is difficult. This paper presents an approach of automated manpower planning model which can be used by MTO operations to achieve a better transparency and synchronization of capacity load for short to medium planning horizons. The approach is implemented as a software tool to automate the data processing and analysis, which helps to dramatically reduce the corresponding data operation efforts and planning time. This paper also presents the validation of the approach and tool in a real production unit in a German small MTO manufacturing company.}, keywords = {information systems, Information visualisation, production planning and control, resource efficient manufacturing}, pubstate = {published}, tppubtype = {article} } In today’s tough economy, it is important for (Make-To-Order) MTO companies to be responsive to customer demand and market fluctuations and to keep the costs as low as possible at the same time. Unlike in Make-To-Stock (MTS), MTO companies hold capacity in reserve. Thus, they are able to make efficient utilization of available capacity to satisfy customer needs. This then leads to a constant capacity planning problem. The companies are facing fluctuations between overload by lack of sufficient capacity, and idleness by excess of capacity comparing to the level of demand. Among all the planning resources, the available manpower is one of the most essential parts of the MTO operations. Therefore, the allocation and adjustment of manpower capacity that suits different planning horizons is a predominant measure to meet the changing capacity demands. Nonetheless, a signing each individual labor to various types of tasks and orders on a day to day basis continually for the planning horizon of several weeks or months is difficult. This paper presents an approach of automated manpower planning model which can be used by MTO operations to achieve a better transparency and synchronization of capacity load for short to medium planning horizons. The approach is implemented as a software tool to automate the data processing and analysis, which helps to dramatically reduce the corresponding data operation efforts and planning time. This paper also presents the validation of the approach and tool in a real production unit in a German small MTO manufacturing company. |
2017 |
McGlinn, Kris; Yuce, Baris; Wicaksono, Hendro; Howell, Shaun; Rezgui, Yacine Usability evaluation of a web-based tool for supporting holistic building energy management Journal Article Automation in Construction, 84 , pp. 154 - 165, 2017. Abstract | Links | BibTeX | Tags: Artificial neural network, BEMS, Fuzzy logic, Genetic algorithm, IFC, Information visualisation, Ontology @article{MCGLINN2017154, title = {Usability evaluation of a web-based tool for supporting holistic building energy management}, author = {Kris McGlinn and Baris Yuce and Hendro Wicaksono and Shaun Howell and Yacine Rezgui}, url = {https://www.sciencedirect.com/science/article/pii/S0926580516303545}, doi = {https://doi.org/10.1016/j.autcon.2017.08.033}, year = {2017}, date = {2017-03-31}, journal = {Automation in Construction}, volume = {84}, pages = {154 - 165}, abstract = {This paper presents the evaluation of the level of usability of an intelligent monitoring and control interface for energy efficient management of public buildings, called BuildVis, which forms part of a Building Energy Management System (BEMS.) The BEMS ‘intelligence’ is derived from an intelligent algorithm component which brings together ANN-GA rule generation, a fuzzy rule selection engine, and a semantic knowledge base. The knowledge base makes use of linked data and an integrated ontology to uplift heterogeneous data sources relevant to building energy consumption. The developed ontology is based upon the Industry Foundation Classes (IFC), which is a Building Information Modelling (BIM) standard and consists of two different types of rule model to control and manage the buildings adaptively. The populated rules are a mix of an intelligent rule generation approach using Artificial Neural Network (ANN) and Genetic Algorithms (GA), and also data mining rules using Decision Tree techniques on historical data. The resulting rules are triggered by the intelligent controller, which processes available sensor measurements in the building. This generates ‘suggestions’ which are presented to the Facility Manager (FM) on the BuildVis web-based interface. BuildVis uses HTML5 innovations to visualise a 3D interactive model of the building that is accessible over a wide range of desktop and mobile platforms. The suggestions are presented on a zone by zone basis, alerting them to potential energy saving actions. As the usability of the system is seen as a key determinate to success, the paper evaluates the level of usability for both a set of technical users and also the FMs for five European buildings, providing analysis and lessons learned from the approach taken.}, keywords = {Artificial neural network, BEMS, Fuzzy logic, Genetic algorithm, IFC, Information visualisation, Ontology}, pubstate = {published}, tppubtype = {article} } This paper presents the evaluation of the level of usability of an intelligent monitoring and control interface for energy efficient management of public buildings, called BuildVis, which forms part of a Building Energy Management System (BEMS.) The BEMS ‘intelligence’ is derived from an intelligent algorithm component which brings together ANN-GA rule generation, a fuzzy rule selection engine, and a semantic knowledge base. The knowledge base makes use of linked data and an integrated ontology to uplift heterogeneous data sources relevant to building energy consumption. The developed ontology is based upon the Industry Foundation Classes (IFC), which is a Building Information Modelling (BIM) standard and consists of two different types of rule model to control and manage the buildings adaptively. The populated rules are a mix of an intelligent rule generation approach using Artificial Neural Network (ANN) and Genetic Algorithms (GA), and also data mining rules using Decision Tree techniques on historical data. The resulting rules are triggered by the intelligent controller, which processes available sensor measurements in the building. This generates ‘suggestions’ which are presented to the Facility Manager (FM) on the BuildVis web-based interface. BuildVis uses HTML5 innovations to visualise a 3D interactive model of the building that is accessible over a wide range of desktop and mobile platforms. The suggestions are presented on a zone by zone basis, alerting them to potential energy saving actions. As the usability of the system is seen as a key determinate to success, the paper evaluates the level of usability for both a set of technical users and also the FMs for five European buildings, providing analysis and lessons learned from the approach taken. |