How Neal Analytics technology can help boost production, quality control

How Neal Analytics technology can help boost production, quality control

A primary challenge for many manufacturers is maintaining quality while also running production fast enough to meet production targets. Achieving that balance requires constant production oversight from human inspectors. To overcome this tradeoff, manufacturers are looking towards autonomous technologies like machine learning, process simulation, and edge computing to help.

Neal Analytics is helping manufacturers address production quality and efficiency challenges with Production Yield Optimization (PYO). Using IoT and edge computing technologies to optimize production processes, PYO extends the capabilities of existing production control systems, human operator capabilities, and improves quality and production throughput. Additionally, PYO incorporates strategies from experienced production employees to control production even when human experts are absent.

How Production Yield Optimization provides efficiency and expertise

PYO brings an intelligent and holistic approach to production processes to improve quality and efficiency. While the solution utilizes advanced autonomous technology, it can work with existing manufacturing equipment.

While traditional control systems can control equipment well when all parameters are stable, PYO can help adjust for variables in everything from raw material characteristics to a facility’s environment. The solution’s integrated machine teaching and deep reinforcement learning capabilities leverage strategies from the supervisors and engineers who know best how to get the most out of an operation’s equipment. Like a human expert, PYO can determine the optimal production settings in order to balance quality and throughput and make sure those settings are maintained through constant monitoring and adjustment.

PYO creates value by transforming subject matter expertise into repeatable control strategies. This autopilot feature can improve a production line’s operation when experienced people aren’t present, and it can free up subject matter experts to focus on higher-value tasks.

The PYO solution is flexible enough to optimize a wide variety of process manufacturing including everything from oil and gas refining to food production. When it came to PepsiCo, Neal Analytics utilized AI to efficiently manufacture Cheetos with consistent flavor and texture qualities while also reducing waste.

For robotic controls in factories and other settings, PYO can automatically adjust for changing configurations and other dynamic conditions. Similarly, our solution can incorporate AI-powered inputs such as computer vision.

How Microsoft and Intel technologies enable production optimization

A number of Microsoft and Intel technologies make various aspects of the Neal Analytics PYO solution excel at production optimization.

Azure cloud technology helps in analyzing data and training computer vision and deep reinforcement learning models. Utilizing the cloud, models can be trained with large datasets and run intensive workloads at scale.

Once trained, PYO is deployed at the edge where Intel modular edge computers optimized for fast inference are used. The key advantage with these edge devices is that they permit low-latency control actions at production facilities. Moreover, the cloud and edge can be used separately or in tandem as needed depending on latency and workload bandwidth requirements.

The autonomous systems used with the PYO solution for extruders are built with the Project Bonsai toolchain and advanced AI and deep reinforcement learning algorithms. These and other advanced tools such as computer vision take advantage of both the cloud and edge as well.

Learn more about Neal Analytics PYO

With our PYO solution, Neal Analytics aims to solve production challenges in the most beneficial way for process manufacturers. Our data scientists and applied engineers understand the complexities that many manufacturers face. Combining our technical knowhow and numerical expertise, we work closely with their subject matter experts to model and solve the right problems from their perspective.

PYO’s integrated machine teaching and deep reinforcement learning capabilities allow the solution to autonomously operate production equipment for peak results. By ensuring end products are produced consistently within specifications for overall quality, the PYO solution can generate significant ROI and help manufacturing operations compete effectively as competitors undergo digital transformations.

Learn more about Neal Analytics Production Yield Optimization on, which showcases intelligent IoT solutions using joint Intel and Microsoft technology. You also can explore other ways Intel and Microsoft are using machine learning and AI to modernize manufacturing.