CGI (Canada)

The accelerated development of the digital economy requires flexible manufacturing systems that can optimize resource planning and decision-making efficiently. Unfortunately, conventional methods are not only inefficient in their scalability but also in making dynamic adjustments in real time. An integrated system introduced in this study employs Holonic Systems (HS) and Latent Dirichlet Allocation (LDA) to enhance manufacturing flexibility, strategic knowledge, and resource handling. The methodology uses holonic architectures for self-organization in production and LDA for topic modeling to support data-driven optimizations. Empirical analysis, by way of mathematical and computational frameworks, measures the performance of systems in terms of accuracy, precision, recall, F1-score, and efficiency in resource management (RME). The outcome shows a high improvement with the RME down to 4.2%, accuracy of 95.6%, and an improved decision-making ability. The research concludes that holonic systems enabled by LDA enhance industrial adaptability and operational efficiency, delivering a sound platform for sustainable digital manufacturing. Future research must further investigate reinforcement learning to enhance predictive decision-making and industrial scalability.

Social media as a class of communication platforms has been investigated in terms of user intentions, or why people use them, in terms of depth or kind of interaction, and in terms of the demographic character of users across various services. There is little understanding however of how and on what grounds people sort the services in which they participate into categories that can explain or predict their patterns of use. One user may sort Twitter.com, Tumblr.com, and Pinterest.com together because of opportunities for interaction, and aggregators like Reddit.com, or StumbleUpon.com because of content. We propose a seed- and-snowball sample that begins with users in two demographic categories, twenty somethings and forty somethings, that asks participants to free pile-sort cards that represent social media services popular in Canada into groups. We will use a short structured interview collected from the initial participants to develop grounds to explain the categorizing criteria the participants demonstrate. We believe the results of the study can explain or predict trends in patterns use, for example the exodus of younger users from Facebook. This is a research work-in-progress though we expect to have results in the form of data and analyses to present at the conference.