China
Africa
Explore chapters and articles related to this topic
Basic Notions of Statistics
Published in Piotr Konieczka, Jacek Namieśnik, Quality Assurance and Quality Control in the Analytical Chemical Laboratory, 2018
Piotr Konieczka, Jacek Namieśnik
Hence, other means have been proposed; for example, the truncated mean. This mean, less sensitive to outliers than the standard mean (only a large number of outliers can significantly influence the truncated mean) and standard deviation, is calculated using all results, which transfers the extreme to an accepted deviation range—thanks to the application of appropriate iterative procedures.
Dynamic asset-liability management problem in a continuous-time model with delay
Published in International Journal of Control, 2022
A. Chunxiang, Yang Shen, Yan Zeng
In the game theoretic framework, the mean-variance ALM problem with delay is formulated as an optimisation problem evaluated dynamically at any t. For each , we solve a truncated mean-variance ALM problem with delay as follows: whose optimal strategy is different from the pre-commitment optimal strategy derived at time 0 in general (see Basak and Chabakauri (2010)). In this case, Bellman's optimal principle is not applicable. Instead, we solve Problem in the game theoretic framework proposed by Björk and Murgoci (2010), and derive an equilibrium strategy of , denoted by , which is rigorously defined in the following definition similar to those proposed by Ekeland and Lazrak (2008) and Ekeland and Pirvu (2008).
Centralised vs. decentralised control decision in card-based control systems: comparing kanban systems and COBACABANA
Published in International Journal of Production Research, 2019
Matthias Thürer, Nuno O. Fernandes, Mark Stevenson, Ting Qu, Cong Dong Li
Operation processing times follow a truncated lognormal distribution (Trietsch et al. 2012) with a truncated mean of 1 time unit and a maximum of four time units. Processing time variability is a factor that is likely to influence the card acquisition time of kanban systems and thus the performance difference between COBACABANA and kanban. Therefore, three levels of processing time variability are modelled, with a squared coefficient of variation, cv2 = 0.25, 0.5, and 1 (after truncation). The level of 1 is equal to the variability of an exponential distribution, which is typically considered to represent ‘high’ processing time variability. The ‘medium’ level of 0.5 is equal to the variability of a two-Erlang distribution. Finally, the level of 0.25 has been chosen to represent ‘low’ processing time variability. This level is still sufficient to avoid unrealistic, nearly symmetric distributions, as observed for lower cv2 levels. Set-up times are considered part of the operation processing time.
Job shop management of products under internal lifespan and external due date
Published in International Journal of Production Research, 2018
Pedro L. Gonzalez-R, Marcos Calle, Jose L. Andrade-Pineda
A simulation model of a randomly routed job shop (Melnyk and Ragatz 1989) has been implemented in Python© using the SimPy© module. This work studies a shop floor composed of eight work stations. As in other works -see Thürer, Silva, and Stevenson (2010), Land (2006) and Oosterman, Land, and Gaalman (2000), the routing length is uniformly distributed between 1 and the total number of stations (eight stations in this study). It will be assumed that a manufacturing order does not visit the same station twice and all stations have an equal probability of being visited. As in recent studies (see Thürer et al. 2016, 2012, and Thürer, Silva, and Stevenson 2011), the processing times follow a truncated 2-Erlang distribution with a non-truncated mean of 1 time units and a maximum of 4 time units. The planned throughput time, Tjs, of job j in station s is set to 5, according to Thürer, Silva, and Stevenson 2010. In this work, set-up times have not been considered.