Challenge: With information from 49 variables collected in real time, identify the costs concerning the occurrence of bones in the chicken breast fillet in the whole process chain, from receiving the live chicken to the cut, economically rank and find the probable sources of issues, including indicating the efficiency of the equipments and the correlation among them.
Solution: BiminD’s solution was installed inside the customer’s factory with access to a dashboard within the headquarters at another state. With the financial, OEE, correlation and simulation modules, the available information are combined and evaluated to pinpoint to the maintenance team the best time for action at the factory, as well as the exact location of the issues. All of this without the need of specialists evaluating the process constantly.
Results: With the implementation, it is possible to reduce the occurrence of quality loss, be more assertive at the maintenances and constantly evaluate if the actions from the maintenance team are being effective, as well as if it is already the best time to train less experienced operators.
Challenge: Identify the best operation point for the burners to achieve less natural gas waste considering a thermal treatment furnace for big steel parts and 20 burners working independently at the same system.
Solution: The control loop evaluation module was installed to understand the optimal operation points versus the interaction between the logics. Afterwards the financial module was utilized to find out the waste of resources (natural gas, stops, personnel, etc.) and convert it into money.
Results: This solution allowed the customer to not only learn with more precision how much the lack of tuning was affecting the consumption of supplies in the factory, but it made possible to identify the best time to make maintenances considering the expected return on investment.
Challenge: Find the best tuning for the process to achieve less consumption of slaked lime and maintain the Ph level of the effluents at a specific value considering both economic factors such as waste in resources and pace of the process. Beyond that, the system should automatically be controlled through BirminD’s solutions.
Solution: The financial, control loops and auto tuning modules were utilized to understand the qualities bottlenecks, identify the optimal operation points, measure the money waste and automatically act to tune the control loops.
Results: Currently the system is capable of self-correction and constantly seek the best tuning to reduce the waste in supplies, resulting in nonstop financial savings. The next steps will consist in evaluate other processes inside the factory and the correlation between these processes, in a way to find probable bottlenecks through the usage of artificial intelligence.
Challenge: Reduce coke consumption at the lime kilns.
Solution: Here the financial and control loops modules were utilized to find the best ratio of consumption x temperature considering the non-linearities of the process.
Results: A improvement of 7.8% was achieved at the coke consumption besides the Discovery of a potential reduction of 700 tons of pollutants in the atmosphere.
Challenge: Determine, among 100+ variables, which ones impact the most the emission of particulate matter of the factory’s chimneys.
Solution: Utilizing the correlation module, machine learning algorithms were applied to find out the most relevant variables for the pollutants emissions.
Results: With this solution it will be possible to learn which variables most affect the pollutants emission and plan prevention methods directly at the source of the problems.
Challenge: Understand the financial effects of the micro-stops and the best operation points of the equipments for a production with more margin.
Solution: The financial, OEE and combination modules were utilized together to create a rank measuring the stops considering money waste, to afterwards learn what are the best operation points of each equipment to obtain the best quality and least waste.
Results: It will be possible to the user to identify what are the main process waste contributors, calculate the return on investment of maintenance services and also learn what are the best operation points to achieve a more efficient and productive industrial plant.