A Reinforcement Learning Control in Hot Stamping for Cycle Time Optimization

Hot stamping is a hot metal forming technology increasingly in demand that produces ultra-high strength parts with complex shapes. A major concern in these systems is how to shorten production times to improve production Key Performance Indicators. In this work, we present a Reinforcement Learning approach that can obtain an optimal behavior strategy for dynamically managing the cycle time in hot stamping to optimize manufacturing production while maintaining the quality of the final product. Results are compared with the business-as-usual cycle time control approach and the optimal solution obtained by the execution of a dynamic programming algorithm. Reinforcement Learning control outperforms the business-as-usual behavior by reducing the cycle time and the total batch time in non-stable temperature phases.

An Internet of Things Platform Based on Microservices and Cloud Paradigms for Livestock

With the growing adoption of the Internet of Things (IoT) technology in the agricultural sector, smart devices are becoming more prevalent. The availability of new, timely, and precise data offers a great opportunity to develop advanced analytical models. Therefore, the platform used to deliver new developments to the final user is a key enabler for adopting IoT technology. This work presents a generic design of a software platform based on the cloud and implemented using microservices to facilitate the use of predictive or prescriptive analytics under different IoT scenarios. Several technologies are combined to comply with the essential features—scalability, portability, interoperability, and usability—that the platform must consider to assist decision-making in agricultural 4.0 contexts. The platform is prepared to integrate new sensor devices, perform data operations, integrate several data sources, transfer complex statistical model developments seamlessly, and provide a user-friendly graphical interface. The proposed software architecture is implemented with open-source technologies and validated in a smart farming scenario. The growth of a batch of pigs at the fattening stage is estimated from the data provided by a level sensor installed in the silo that stores the feed from which the animals are fed. With this application, we demonstrate how farmers can monitor the weight distribution and receive alarms when high deviations happen.

Using PRRSV-Resilient Sows Improve Performance in Endemic Infected Farms with Recurrent Outbreaks

Simple Summary

Porcine reproductive and respiratory syndrome (PRRS) is a viral disease responsible for huge economic losses to the pig industry. The selection of PRRSV resilient sows has been proposed as a strategy to control this disease. A simulation model was developed to test the differences in reproductive performance and economic outcome of resilient or susceptible sows under farm PRRSV endemic conditions with or without recurrent PRRSV outbreaks. The data from phenotyped sows came from a PRRSV-positive farm with 1500 sows that suffered a PRRSV outbreak that lasted 24 weeks within three years. The reproductive parameters were generally better for resilient than for susceptible sows in PRRSV-positive farms suffering recurrent PRRSV outbreaks. Consequently, the piglet production cost was lower for resilient than for susceptible sows in any condition but showed only significant differences in PRRSV endemic farms suffering recurrent outbreaks. Finally, the annual gross margin by sow is significantly better for resilient than for susceptible sows under endemic conditions with or without recurrent outbreaks. Thus, the selection of PRRSV resilient sows is always a profitable approach for producers supporting the control of this disease.

Abstract

The selection of porcine reproductive and respiratory syndrome (PRRS) resilient sows has been proposed as a strategy to control this disease. A discrete event-based simulation model was developed to mimic the outcome of farms with resilient or susceptible sows suffering recurrent PRRSV outbreaks. Records of both phenotypes were registered in a PRRSV-positive farm of 1500 sows during three years. The information was split in the whole period of observation to include a PRRSV outbreak that lasted 24 weeks (endemic/epidemic or En/Ep) or only the endemic phase (En). Twenty simulations were modeled for each farm: Resilient/En, Resilient/En_Ep, Susceptible/En, and Susceptible/En_Ep during twelve years and analyzed for the productive performance and economic outcome, using reference values. The reproductive parameters were generally better for resilient than for susceptible sows in the PRRSV En/Ep scenario, and the contrary was observed in the endemic case. The piglet production cost was always lower for resilient than for susceptible sows but showed only significant differences in the PRRSV En/Ep scenario. Finally, the annual gross margin by sow is significantly better for resilient than for susceptible sows for the PRRSV endemic (12%) and endemic/epidemic scenarios (17%). Thus, the selection of PRRSV resilient sows is a profitable approach for producers to improve disease control.