Real-time identification of mean retention time and end-of-life rate in cyclic manufacturing

Used in predicting demand and new orders for shelf stock replacement and in inventory planning, the mean retention time (MRT) of consumer products is long established as a key production and marketing parameter. Higher MRTs are also associated with enhanced sustainability, favoring longer production/consumption cycles, waste prevention and resource preservation. The MRT, or the corresponding product stock level and the end-of-life (EoL) outflow to be collected for disassembly and recycling are useful for planning efficient cyclic operations. Determination of stock and EoL level for cyclic products is challenging however, due to random stock losses and to the volatile and distributed nature of returns and EoL exit. Actuary science methods presume specific residual life distributions, usually fitted ex-post, which may become inaccurate when applied to real production/consumption cycles. Anchored on a recent constitutive law in cyclic manufacturing, a method is presented for real-time MRT and EoL identification. The law related mean stock and EoL levels under random losses and random, distributed EoL exit via a simple equation, without involving residual life or EoL distributions or early loss history. One degree of freedom still remained, for, both stock and EoL are random and unobservable. The work herein ties the remaining degree of freedom. It enables sequential identification of stock and EoL level, using reliable and readily acquired data in a key equation obtained for the production/consumption cycle. Adapting to varying early loss, consumer take-back and discard volatility, it may be useful in proactive policy, sizing of disassembly and recycle lines and return inventory control towards cleaner production/consumption cycles.