Software : Data & Analytics : IIoT
Hitachi provides solutions to customers in a range of sectors, including power/energy, industry/distribution/water and others. The Company operates in eight segments. The Information & Telecommunication Systems segment provides system integration, consulting, cloud service and others. The Social & Industrial Systems segment provides industrial equipment and plants, thermal power, nuclear power and natural energy power generation systems and others. The Electronic Systems & Equipment segment provides semiconductor manufacturing equipment and others. The Construction Machinery segment provides hydraulic excavators, wheel loaders and others. The High Functional Materials segment provides materials for semiconductors and displays and others. The Automotive Systems segment provides engine powertrain systems and others. The Smart Life & Ecofriendly Systems segment provides business air conditioners and others. The Others segment provides optical disk drive products and others.
How can we help reduce plastic waste? Facilitating the use of recycled plastics using in-mold sensors to optimize the injection molding process
To use recycled materials with material properties that fluctuate from lot-to-lot, conventionally, a skilled operator made fine manual adjustments to the injection molding process conditions according to the material properties. As this is time consuming and requires experience, this has limited the type and amount of recycled materials used as manufacturers have sought to use recycled materials with consistent properties.
To address this issue, my colleagues and I conducted a study where we looked at how we could automatically optimize the process conditions and thereby contribute to quality, and presented our results at the 37th International Conference of the Polymer Processing Society (PPS-37) which was held in Fukuoka, Japan, from 11-15 April 2022. Below, I’d like to briefly share what we did.
Application of deep learning methods for more efficient water demand forecasting
In recent years, such predictions have also found wide application in near-optimal control operations of water networks. Water demand prediction is an active field, where different methods and techniques have been applied including conventional statistical methods and machine learning methods. Due to advancements in the field of sensing and IoT, an increasing amount of data is becoming available for water distribution systems, including water demand data. Therefore, we are seeing greater use of deep learning methods to develop models for water demand forecasting in recent years as deep learning methods can deal with seasonality as well as random patterns in the data, and provide accurate results compared to traditional methods.
We observed that the frequency of data, amount of data, and quality of data has an impact on the deep learning model accuracy. In CNN-LSTM, CNN effectively extracts the inherent characteristics of historical water consumption data such as seasonality, and LSTM can fully reflect the long-term historical process and future trend. Hence, water demand forecast predictions using CNN-LSTM produced a better result when compared to other single models such as GRU, MLP, CNN and LSTM.
Manufacturing line design configuration with optimized resource groups
Skilled line engineers spend several months designing a manufacturing line based on their experience. Optimization of the four design specifications from the viewpoint of productivity and equipment continuity is required for the line design process. However, these four design specifications are highly dependent on each other and the number of feasible combinations of the specifications is enormous and difficult to automate.
To solve these issues, our research introduces the concept of a resource group that enables a methodology to solve the four design items hierarchically and develops methods to quickly build new manufacturing lines in response to changes in product varieties and manufacturing fluctuations in a factory.
Operation planning method using convolutional neural network for combined heat and power system
The energy efficiency of a combined heat and power (CHP) can reach about 85%, whereas conventional thermal power plants operate only at 45% efficiency or lower. CHPs perform better mainly because the heat from generators can be used as a energy source to meet heat demands or power refrigerators to generate cold water, in other words the “waste” heat is used and not wasted. Therefore, a growing number of factories and commercial buildings are installing combined heat and power (CHP) systems that include various energy storage devices. To reduce the energy cost of CHPs, optimal operation plans to satisfy time-varying energy demands with minimum energy cost are required. However, conventional operation planning methods using optimized calculation have an issue with long computing time. Especially these days, operation plans need to be generated within a few minutes or even seconds to make up for output of renewable energy sources.
Action-limited, multimodal deep Q learning for AGV fleet route planning
In traditional operating models, a navigation system completes all calculations i.e., the shortest path planning in a static environment, before the AGVs start moving. However, due to constant incoming offers, changes in vehicle availability, etc., this creates a huge and intractable optimization problem. Meanwhile, an optimal navigation strategy for an AGV fleet cannot be achieved if it fails to consider the fleet and delivery situation in real-time. Such dynamic route planning is more realistic and must have the ability to autonomously learn the complex environments. Deep Q network (DQN), that inherits the capabilities of deep learning and reinforcement learning, provides a framework that is well prepared to make decisions for discrete motion sequence problems.
A combinatorial auctions approach to capacity sharing in collaborative supply chains
Collaboration among multiple parties throughout the supply chain may improve resilience. In collaborative supply chains bilateral agreements between buyer and supplier are replaced or supplemented by multilateral agreements. Three major complications for multi-party collaboration in automotive supply chains arise. Existing bilateral contracts must be respected, procurement of parts needs to be bundled, and the delivery of parts is constrained temporally, to ensure available production capacity and minimise inventory.
Using digital twin for cost-efficient wind turbines
CBM of the wind turbine is usually conducted by monitoring vibration at many points on each component with dedicated sensors. Simply increasing the number of monitored points and components leads to an increase in monitoring cost. In our approach, the digital twin acts as virtual sensors for monitoring any component whose behavior can be simulated from a smaller number of sensors as input to the digital twin. Thus, CBM with the digital twin contributes to identifying critical turbines, components, and positions that need maintenance.
Medicine piece-picking robot for Hitachi Transport System
Applying deep learning to sensor data to support workers in manufacturing
To achieve next-generation production systems and Multiverse Mediation with CPSs, 4M (huMan, Machine, Material, and Method) work transitions need to be clarified and used more accurately. However, traditional systems cannot detect deviations in manual procedures. To resolve these issues, we are developing a highly accurate detection technology for “human work”. Figure 2 shows the assembly cells considered in this study.
Compared to conventional approaches, we achieved a 15% reduction in product assembly time and a deviation detection leak of almost zero (more than 95% work identification accuracy). These results demonstrated the potential for our system to efficiently and effectively support manufacturing workers and contribute to greater efficiency and quality management in the assembly of complex equipment.
Fabs Drive Deeper Into Machine Learning
For the past couple decades, semiconductor manufacturers have relied on computer vision, which is one of the earliest applications of machine learning in semiconductor manufacturing. Referred to as Automated Optical Inspection (AOI), these systems use signal processing algorithms to identify macro and micro physical deformations.
Defect detection provides a feedback loop for fab processing steps. Wafer test results produce bin maps (good or bad die), which also can be analyzed as images. Their data granularity is significantly larger than the pixelated data from an optical inspection tool. Yet test results from wafer maps can match the splatters generated during lithography and scratches produced from handling that AOI systems can miss. Thus, wafer test maps give useful feedback to the fab.
Smart operation recommender system digitalizing OT knowledge to improve productivity
To contribute to resilient manufacturing systems through digitalization, it will be important to connect physical phenomena, OT knowledge and sensor data in parallel with developing AI to analyze and learn from the data. My colleagues and I developed an operation recommender system that can discriminate factors contributing to defects and recommend appropriate countermeasures. We will continue developing the technology to resolve issues in manufacturing by considering different manufacturing processes.
Digital twin for load monitoring of wind turbine blade
Recently, the lifetime extension of wind turbines has increasingly attracted attention as one way to reduce levelized cost of energy. To explain, generally, wind turbines are designed under the wind condition defined by design standards such as the International Electrotechnical Commission (IEC), however, real wind conditions do not always correspond to the design condition. Therefore, the actual lifetime of wind turbines can be extended when the real wind condition is less severe than the design condition. For the lifetime extension, however, it is important to have an accurate evaluation of remaining useful lifetime (RUL). To accurately evaluate RUL, we should know historical data of loads applied to a structure of wind turbine but unfortunately, often there are not enough sensors to provide a full set of data to evaluate the loads. Thus, while the simple solution would be to add more sensors for the load evaluation, this would defeat the purpose as it would entail additional costs, and thus reduce the goal of trying to reduce the levelized cost of energy through lifetime extension. So, the challenge is to accurately estimate the load from the sensor data available.
Influence estimation for generative adversarial networks
Expanding applications [1, 2] of generative adversarial networks (GANs) makes improving the generative performance of models increasingly crucial. An effective approach to improve machine learning models is to identify training instances that “harm” the model’s performance. Recent studies [3, 4] replaced traditional manual screening of a dataset with “influence estimation.” They evaluated the harmfulness of a training instance based on how the performance is expected to change when the instance is removed from the dataset. An example of a harmful instance is a wrongly labeled instance (e.g., a “dog” image labeled as a “cat”). Influence estimation judges this “cat labeled dog image” as a harmful instance when the removal of “cat labeled dog image” is predicted to improve the performance (Figure 1)
AI In Inspection, Metrology, And Test
“The human eye can see things that no amount of machine learning can,” said Subodh Kulkarni, CEO of CyberOptics. “That’s where some of the sophistication is starting to happen now. Our current systems use a primitive kind of AI technology. Once you look at the image, you can see a problem. And our AI machine doesn’t see that. But then you go to the deep learning kind of algorithms, where you have very serious Ph.D.-level people programming one algorithm for a week, and they can detect all those things. But it takes them a week to program those things, which today is not practical.”
That’s beginning to change. “We’re seeing faster deep-learning algorithms that can be more easily programmed,” Kulkarni said. “But the defects also are getting harder to catch by a machine, so there is still a gap. The biggest bang for the buck is not going to come from improving cameras or projectors or any of the equipment that we use to generate optical images. It’s going to be interpreting optical images.”
Is maintenance worth the money? A data-driven answer
For many industrial and commercial operations, maintenance accounts for a large part of operating costs. For instance, maintenance costs range from 15% to 60% of the total production costs in manufacturing plants. In the airline industry, the 2014 global spend on maintenance, repair, and overhaul accounted for around 9% of the total operational costs, and this spend is estimated to reach 90 billion dollars in 2024. Even with maintenance cost being such a substantial part of the overall cost, maintenance managers have little visibility into whether maintenance expenditure is money well spent or not, i.e., whether the maintenance is effective or not. In this blog, I’d like to talk about a mathematical formulation of the maintenance effectiveness evaluation problem and a systematic way of solving it.
Knowledge base construction to improve voice-enabled AI in industrial settings
In this research, we focus on building a framework for constructing a KB of equipment components and their problems entities with “component
Equipment Health Indicator Learning using Deep Reinforcement Learning
We propose a machine learning based method to solve health indicator learning problem. Our key insight is that HIL can be modeled as a credit assignment problem which can then be solved using Deep Reinforcement Learning (DRL). The life of equipment can be thought as a series of state transitions from a state that is healthy at the beginning to a state that is completely unhealthy when it fails. Reinforcement learning learns from failures by naturally backpropagating the credit of failures into intermediate states.