Big data analysis for the estimation of disassembly time and de-manufacturing activity

Design for disassembly is a key enabling strategy for the development of new business models based on the Industry 4.0 and circular economy paradigms. This paper attempts to define a method, based on Data Mining, for modelling disassembly data from large amount of records collected through the observation of de-manufacturing activities. The method allows to build a repository to characterize the disassembly time of joining elements (e.g. screws, nuts) considering different features and conditions. The approach was preliminary tested on a sample of 344 records for nuts disassembly retrieved by in-house tests. Disassembly time and corrective factors were assessed including the analysis of probability distribution function and standard deviation for each feature (i.e. disassembly tool).