Order-related acoustic characterization of production data
- PDF / 1,041,432 Bytes
- 10 Pages / 595.276 x 790.866 pts Page_size
- 84 Downloads / 188 Views
ORIGINAL PAPER
Order-related acoustic characterization of production data M. Iber • K. Windt
Received: 20 August 2012 / Accepted: 23 August 2012 / Published online: 5 September 2012 Ó Springer-Verlag 2012
Abstract The conductor of an orchestra is able to distinguish not only between different instruments, but even among dozens of string players performing on instruments with similar sound qualities. Trained human ear not only is capable to highly differentiate between pitches and colors of sound, but also to localize the position, where the sound is coming from. This paper presents a parameter mapping sonification approach on production data, which is based on these human perceptual skills. Representatively for other logistic parameters, throughput times of orders are sonified and allocated in a sonic space. Additionally to auditory representations of the established resource and order oriented views in logistics, a third perspective is introduced, which displays the complete workflow of an order simultaneously as a multi-pitched spatial sound. Thus, causes and impacts of high throughput times in the data set example could be identified. Keywords Manufacturing Parameter mapping sonification Data mining Logistic analysis
1 Introduction Profound analysis of actual and planning data and their correlation is an essential requirement for the adjustment of operating levers in production planning and control. Depending on the amount of work systems of a production M. Iber (&) K. Windt Jacobs University Bremen, Campus Ring 1, 28759 Bremen, Germany e-mail: [email protected] K. Windt e-mail: [email protected]
shop, the number of product variants to be produced, and the quantity of restrictions caused by technical requirements or customer demands, the structure of manufacturing data easily reaches the complexity of NP-hard problems [1]. Whereas traditional methods [2] rely on averaging in order to reduce complexity, more recent approaches include advanced statistics and data mining [3] for a deeper understanding of production data. An important component of both, data-mining and traditional statistic approaches as applied in logistic analysis, is exploratory data analysis (EDA). The term [4] comprises the participation of a human analyzer, who interactively explores the structure of data in recursive proceedings between generating and proving hypotheses. Well-established approaches are graphical statistics and data visualizations [5]. In the context of chronologically structured data such as production data, the acoustic equivalent to graphical display, the auditory display of statistical data (as provided by sonifications) is a promising method to gain knowledge about temporal fluctuations of bottlenecks in production workflows. In natural science, auditory display still is widely disdained in comparison with its visual correspondent [6]. This might be caused by the visual alignment of human thinking per se including written language as the legitimate form to capture thoughts and scientific re
Data Loading...
