Seeq and AspenTech accelerate self-service industrial analytics on AWS

Introduction

Seeq, a global leader in advanced analytics for process manufacturing industries, uses Amazon Web Services (AWS) to power its software-as-a-service (SaaS) platform for industrial time-series analytics. Seeq helps process experts, equipment experts, and data scientists perform advanced analytics for industry at scale. Drive your sustainability and operational excellence goals faster and more easily than ever before, across your enterprise. With Seeq SaaS on AWS, you can empower your subject matter experts (SMEs) to generate near real-time (NRT) insights into the health of your assets and processes, in addition to doing root-cause analysis, optimizing energy usage, reducing downtime, and more.

Aspen InfoPlus.21 (IP.21) is an industrial process historian-designed to capture and manage vast volumes of time-series data. With its scalable architecture, IP.21 provides a unified view of operations, whether for a single facility or a global network of plants. Offering NRT visibility, IP.21 helps companies closely monitor assets and facilitates the prompt identification and correction of inefficiencies. Its advanced analytical tools cover trend analysis, anomaly detection, and predictive insights. Seamlessly integrating with platforms like Seeq and AWS, IP.21 delivers smooth data flow and optimized decision-making. In essence, IP.21 is a strategic tool that drives operational excellence in industry. Seeq customers have already connected over 3 million IP.21 signals to monitor more than 400,000 assets and processes.

In this blog, you will learn how Seeq on AWS taps into isolated operational technology (OT) data collected in IP.21 and helps leaders in process engineering and data scientists develop insights, optimize processes, and enhance business value.

What use cases/industry pain points are covered by Seeq + IP.21?

Every expert is an analyst
Seeq accelerates industrial analytic workflows by giving SMEs access to IP.21 and connecting them with relevant data—on demand and virtually direct from the source. Seeq has a no-code/low-code environment, interactive dashboards and reporting with drilldown, and a JupyterLab environment that is ideal for Python power users, and having these different interface options means that users of different profiles can analyze and experiment with data. For example, you can power downstream AWS workloads, such as machine learning (ML) applications, with SME-enriched, labeled data or drive rapid returns on infrastructure investments with analysis of OT data and insights into performance KPIs obtained by your SMEs.

Asset and process performance monitoring
Operators, engineers, and managers require NRT monitoring of processes and asset performance. Seeq, a solution by industry experts for industry experts, is purpose-built for industrial time-series data. Seeq queries data from IP.21 and performs an SME’s calculation workflow either on demand or according to a schedule. With Seeq’s semantic layer asset hierarchy, organizations can effortlessly scale calculations across various assets and sites, facilitating uniformity and efficiency in monitoring. Incorporate Seeq and IP.21 data into risk-based maintenance planning based on expert analyses, and compare current equipment and process operating profiles to historical records or create monitoring regimes by exception. By strategically planning maintenance outages, engineers can not only minimize unplanned downtime but also proactively prevent potential single failure events, thereby sustaining operational excellence.

Production optimization
Seeq can combine both time-series and event data in analyses, a capability that is often critical to customers with batch operations. Identify batch phases and subphases through analysis of process operating data by combining step numbers, flow rates, and valve modes. Capsules can identify different process phases and calculate key metrics for each one, resulting in batch-phase and long-term views of recommended asset and process improvements, potentially saving time and increasing efficiency. Seeq users have access to continual contextualization of the process data stored in IP.21, helping process engineers create accurate process signal statistics for each mode of operation (for instance, operators can opt to totalize flow rates only when in cleaning mode rather than in sterilization mode or identify max conductivity per cleaning event). Additionally, by providing a central view into operational data, Seeq streamlines the process of documenting processes and sharing insights, accelerating the identification and resolution of operational problems.

Balancing IT and OT strategies
Chief digital officers (CDOs) and other C-level executives across the industrial sector are striving to balance IT (information technology) and OT (operational technology) strategies while navigating the business transformation inherent in digitization initiatives. Business imperatives can be grouped into three categories: holistic risk management in volatile and uncertain markets, margin improvement in challenging environments, and sustainability goals in conjunction with profitability and productivity targets. The question quickly becomes, Which of these categories are IT and which OT imperatives? A recurring observation is that many customers grapple with the multifaceted challenge of accurately defining the scope of a given problem and pinpointing the optimal blend of talent, technology, and process to address it. Integrating IT strategies with OT systems like IP.21 facilitates data-driven decision-making that is based on a rich history of operational data, leading to more informed and strategic choices.

Build OT data lakes on Amazon S3
Engineers spend immense amounts of time and effort cleaning and transforming the OT data locked in relevant systems before it can be analyzed. The inherent inconsistencies in OT data, such as varying identifiers, anomalies, and communication glitches, pose significant hurdles. As a result, process engineers can find themselves investing months in data cleaning and preparation before diving into meaningful analysis. With Seeq’s connector to IP.21—powered by AWS Glue, which helps companies discover, prepare, and integrate all their data at any scale—you can reduce the time it takes to clean and transform your data from weeks to hours. This streamlined process culminates in the refined data being stored in an Amazon S3 data lake, where it is ready to be used for in-depth analysis. The benefits of this integration are many, with use cases including the following:

• Revealing insights from data to reduce operations costs, improve resource utilization, and lower the risks associated with business operations
• Supporting business decisions with data from globally distributed operations and systems
• Implementing predictive analysis to deliver actionable insights across your organization
• Performing fleet-wide analysis to compare asset performance

Make the most of machine learning on AWS
Data scientists can use cleaned process manufacturing data from IP.21 in Seeq and create ML models in Amazon SageMaker, a service for building, training, and deploying ML models for any use case. Those models can be accessed in Seeq, facilitating virtually seamless collaboration between data scientists and SMEs. Seeq also integrates with AWS for Industrial services like Amazon Lookout for Equipment, a service for avoiding unplanned downtime, and supports Internet of Things (IoT) data from Amazon IoT SiteWise, a service for collecting, organizing, and analyzing data from industrial equipment at scale, as well as Amazon Kinesis, a service for collecting, processing, and analyzing NRT video and data streams.

Solution: Integrating AWS, Seeq, and IP.21 for enhanced data analysis

The architecture diagram below shows a virtually seamless hybrid deployment. In this arrangement, IP.21 operates on premises while the Seeq platform is hosted on AWS. Communication between the on-premises IP.21 infrastructure and the Seeq platform is facilitated by the Seeq remote agent. As data travels between devices and networks, secure communication protocols and encryption techniques are used to prevent interception or tampering by malicious actors. Transport Layer Security (TLS) and Secure Sockets Layer (SSL) protocols are commonly used to establish secure connections, helping keep data confidential during transmission. Additionally, strong authentication mechanisms and access controls further fortify the overall data security posture, with only authorized http clients having access to sensitive information.

Figure 1. Architecture diagram

1. First, the IP.21 historian is running on premises, such as an industrial plant or a factory, and captures detailed time-series data alongside relational and event databases. Using the Seeq remote agent, you can access this high-fidelity time-series data without the need for extract, transform, load (ETL) processes. There is native connectivity to IP.21 and a Java Database Connectivity (JDBC) API for connection to databases of different types, including relational, event, laboratory information management system (LIMS), computerized maintenance management software (CMMS), and manufacturing execution system (MES) databases.
2. The Seeq remote agent provides ad hoc data access to cloud data sources with connectivity to various AWS services, including Amazon Athena, a service for analyzing petabyte-scale data, Amazon Redshift, a service for powering data-driven decisions with the best price-performance cloud data warehouse, and Amazon Relational Database Service (Amazon RDS), a service for setting up, operating, and scaling a relational database in the cloud with just a few clicks. Connect to IoT sensor data streamed to Amazon Timestream, a fast, scalable, and serverless time-series database, via Amazon Kinesis Data Streams, a service for easily streaming data at any scale.
3. Process and equipment experts can explore all connected data sources from Seeq’s no-code/low-code self-service analytics solution with Seeq SaaS on AWS. Seeq SaaS provisions secure, dedicated, isolated instances of Seeq for each customer, with no shared databases or storage. Collaborate with colleagues by documenting analysis steps for others to follow or fork. Scale across assets or sites with a single click, using synchronized asset trees or custom asset groups. Interactive dashboards and reporting link back to the original analysis with drilldown. Power users and data scientists can create advanced scripted analysis in a JupyterLabs Python environment and scale those routines across all users through add-ons for the no-code/low-code application.
4. SMEs contextualize, enrich, and label data for downstream advanced workloads like ML. The Seeq Python (SPy) Module gives read/write access to Seeq from any cloud service supporting Python. Access labeled training data from Seeq in Amazon Sagemaker for data science or advanced statistical analysis. Read raw data, prepared data, and analytic results into AWS Glue. Query current process values from Seeq for model endpoint calculations using Amazon Sagemaker or AWS Lambda, a service for running code without thinking about servers or clusters. SMEs can deploy and consume Amazon Lookout for Equipment anomaly detection from within the no-code/low-code application with no programming experience and full integration.
5. Write advanced workload model output and inference calculations back to AWS services like Amazon S3, object storage built to retrieve any amount of data from anywhere, Amazon RDS, or Amazon Redshift, closing the loop between OT SMEs and data scientists.
6. Seeq users can access raw data and advanced model output calculations within the same familiar environment. Operationalize SME analysis alongside ML analysis for process optimization, asset health, and sustainability insights. Report out consolidated analyses to the organization for improved decision-making. Provide OT expert feedback on ML models—in the same context as the OT data that generated the training set. Rapidly experiment and iterate across virtually any use case for virtually any combination of connected data sources.

Conclusion

With Seeq powered by the wealth of data stored in IP.21 running on AWS, you can clean, perform calculations on, and analyze IP.21 data—including context from relational data sources such as MES, batch, and other applications—to diagnose and predict issues and share findings across the organization. With NRT expert collaboration and deeper insights, Seeq helps organizations advance toward their sustainability and operational excellence goals. By tapping into rich data from IP.21, Seeq helps substantially reduce maintenance costs and minimize downtime. You can set up advanced workflows like ML with data-driven, state-of-the-art methods already proven in critical industries using the Seeq SaaS platform in conjunction with the AWS Cloud. The Seeq SaaS solution is listed on AWS Marketplace, making it easier to procure, deploy, and manage your workload. For more detail on pricing for customers in process industries, contact Seeq at info@seeq.com for a private offer.