Bob Factotum is an intelligent tool that allows you to use Artificial Intelligence to derive production information and data, including those about machinery, more simply and intuitively.
Bob Factotum connects to the ERP software and machinery (compliant with Industry 4.0) and allows for natural language interaction thanks to the identification of meta tags in the dataset.
Bob Factotum is based on the conception that the added value given by the end user is about data interpretation rather than the creation of reports using complex User Interfaces (UIs).
Furthermore, Bob Factotum comes with an advanced neural network that allows it to make previsions that are more accurate than the ones made by the Accounting Software using traditional statistical tools. It is possible to use these forecasts in two ways:
– Asking them directly to Bob: “Assuming that the Gamma machinery will not be operating due to scheduled maintenance, how should we re-organize production for the next week?”
– Using a proactivity function, which allows Bob to autonomously conduct simulations on data and identify potential opportunities and risks: “The Gamma machinery needs maintenance since its sensor notify anomalies that may lead to a failure in the next 12 days with a chance of 78%.”
What it does
Understand natural language to interact with the company’s data
Bob Factotum is capable of understanding natural language and convert it into an input to communicate with the ERP software and machinery. The use of natural language as a mean of interaction removes the learning curve barrier, thus saving significant amounts of time and having the user focus only on the activities that create value.
Creation of forecasts about future scenarios using “White Box” Neural Networks and causal correlation
Bob Factotum can create future scenario analyses thanks to the AI networks developed by Divisible Global, more potent than the traditional prevision systems used by ERP software.
Divisible’s causal correlation algorithm is capable of identifying cause-effect relations within data (and not only mathematical ones, as happens with statistical correlation), while the White Box neural network provides a detailed overview of the most relevant variables that affect the end result.
Using proactivity to autonomously produce scenario simulations to identify potential opportunities and risks
Bob Factotum can autonomously simulate scenarios to immediately identify opportunities and risks that may happen thanks to its Proactivity function, all with settable custom goals that can act as guidelines for Bob’s analyses on the goodness of scenarios. For example, if we want to reduce the incidence of set-up costs, Bob can conduct dynamic allocation forecasts for production to minimize the time in which machinery does not operate.
Supervised and Unsupervised Learning
Bob Factotum can learn to improve its predictive model as time passes, by comparing old forecasted data and actual data (supervised learning) and analyzing new data (unsupervised learning).
Customization of models through Hornet AI
Bob Factotum is capable of developing predictive models that answer to the primary needs of a company and is the result of years of field studies and experiences. But, since we know that every firm has different needs and preferences, we developed an integration with our Hornet AI technology for Bob Factotum that allows for the independent creation of machine learning models to answer to all the question that a company may have.
Data Storytelling: Optimized report creation
Bob Factotum can help the user in transforming the multitude of data into useful information to take effective decisions – one of the hardest challenges nowadays. Bob Factotum can recognize data types and guide the user in the creation of relevant and solid reports.