Selecting a process manufacturing analytics solution: five questions to ask
The large gap between data and insight in process manufacturing will only start closing when data analytics vendors start putting the process engineer first.
Process manufacturing organisations run on data — from a manufacturing, operations and business perspective.
The data generation and collection strategies at the centre of manufacturing processes have evolved dramatically, especially in recent years. Process manufacturers now collect and store huge volumes of data throughout their operations, both on and off premise, across multiple geographic locations, in an increasing number of separate data silos.
These advances have coincided with the proliferation of connected sensors and increasingly inexpensive storage, leading to an Industrial Internet of Things (IIoT) projected to generate more than 4 trillion gigabytes of data per year by 2020, according to IDC Research.
Data analytics solutions have a huge positive impact on the growing volumes of data in many sectors, from retail to finance. So why aren’t these solutions widely leveraged in process manufacturing? With so much data and the promise of so many new technologies, why is it so difficult to apply these technologies to process manufacturing and gain the same benefits as other sectors?
Why do process manufacturing organisations still feel like they have too much data and too little insight?
This gap — between the data process manufacturing organisations have and the insights achieved — exists because some data analytics solutions fail to completely grasp the unique challenges and opportunities presented by process manufacturing. Aggregating data from different sources (eg, process historians, sensors) is especially challenging for analytics solutions that were not developed specifically for process manufacturing.
When we talk about data analytics solutions, we mean any software enabling process engineers or scientists to:
- create a cleansed, focused data set for analysis through assembling, aggregating or wrangling data from various sources, including data historians, offline data, manufacturing systems and relational databases;
- investigate operations data using ‘self-service’ tools to rapidly analyse alarm, process or asset data for ad hoc or regular reporting;
- publish or share insights and reports across the organisation to enable data-driven action or enable predictive analytics on incoming data.
Many data analytics solutions claim to offer some or all of these things — with the goal of finally closing the gap between data and insight. But are they successful, and what are the criteria that determine success?
In this article are five questions every process manufacturing buyer should ask when evaluating a data analytics solution.
1. Is the solution designed specifically for process manufacturing, and can it handle time-series data and solve intricate process manufacturing problems?
Why it matters
Anyone who works with process manufacturing data knows it isn’t like other data. No matter the industry — from pharmaceutical to mining to oil and gas — the data produced and the assets involved present a tangle of convoluted relationships and contextual challenges.
Whether you’re looking at a refinery, a production line or a wind turbine farm, there are historians collecting data across many different protocols used by multiple vendors across a disparate array of equipment, all of varying ages and implementations.
These systems are typically producing data at speeds and volumes that other industries would find dizzying, and at uneven intervals that can confound conventional analytics solutions. All this data also needs to be cleansed to be useful.
To make matters worse, all these events and signals lack the associated context to make them meaningful on their own — a problem that is further compounded when assembling data from multiple sources which requires the addition of these key relationships.
Finally, process manufacturing data is hard to navigate. Sensors have timestamps that need to be aligned and aggregated across specific ranges in time. It is difficult to get answers to even the simplest of questions and overcome hurdles that transaction data doesn’t have.
What is needed
Many analytics, reporting and industrial control systems can present data, but process manufacturers have far fewer options for solutions that can coordinate and aggregate data in helpful ways. In your evaluation, you should consider how easily the analytics solution makes it for users to select an appropriate time series for analysis and synchronise input from various control systems, sensors and other sources (which may report at different intervals) to present an accurate view of activity within the specified time window.
An analytics solution should do the heavy lifting of synchronising time-series data from different sources so your specialists have a clear, accurate data set to work from. The solution should not require them to download data from different systems (control systems, databases, sensors, spreadsheets etc) and coordinate it by hand in spreadsheets before any meaningful analysis can begin. It should also support the most useful analysis and output formats for time series data, such as pattern search, value search and scatterplot charts.
2. Does the analytics solution rely on your experts or its experts?
Why it matters
Beware of experts bearing correlations. Many data analytics vendors know their own technology extremely well, but don’t know much about process manufacturing. For example, an analytics solution might tell an oilfield operator that production is diminishing at a specific well, but would not point the operator to check a specific asset, which an industry-focused solution could do. Lack of process or vertical expertise can lead to a focus on the analytics themselves rather than the implications of any findings — and, in turn, an emphasis on correlations over outcomes.
The key to positive business outcomes for process manufacturers is empowering subject-matter expert employees. A typical process manufacturing organisation has a great deal of expertise at its disposal, spread out across different roles including skilled process engineers, analysts, architects, team leads and other technical specialists.
These experienced frontline users often have decades of experience, detailed knowledge of the company’s processes and history, situational awareness of its operations and fluency in plant assets, sensors and tags. They have the advanced technical education and experience to ask smart, productive questions.
Unfortunately, these employees are often limited by an ageing suite of software tools, most of which were originally created in the mid- to late-1990s. They know the right questions to ask — but using existing data analytics tools to answer those question can be difficult and time-consuming, approaching impossible in many cases.
What is needed
Packaged analytics solutions won’t know more than the process engineers and quality specialists that focus on your operations every day, and they shouldn’t pretend to. Instead, data analytics solutions should make it easy for your specialists to access and work with the data they need so they can apply their particular expertise.
Process engineers excel at solving problems and driving incremental improvements by studying how slight changes impact performance. Solutions that save time by helping engineers isolate variables will return tremendous value because users can focus on making changes and modelling results, instead of collecting, coordinating and preparing data.
Analytics solutions for process manufacturing should put power into the hands of the people who can most efficiently create the most positive outcomes by providing productivity tools and features to help them assemble, cleanse, search, visualise, contextualise, investigate and share insights from process data — all without involving IT experts.
If you give the right tools to the right people, you can immediately see positive results.
3. Is the analytics vendor more focused on the problems being solved or the technologies involved?
Why it matters
We’re in the midst of an overwhelming wave of innovation that includes software innovation with big data, hardware innovation highlighted by highly scalable, on-demand computing architectures and cognitive computing innovation marked by ongoing advances in machine learning. These rapid advancements have led to two big problems for anyone trying to compare analytics solutions for process manufacturing organisations.
First, emerging technologies — big data, predictive analytics, machine learning, cloud computing and others — have eclipsed the narrative of benefit and impact. The capabilities of these technologies have surpassed our understanding of the best way to get value from them. Rather than discussing why we should adopt a particular innovation, the conversation focuses too often on what technology to use, often with more enthusiasm for the technology than the actual benefits.
Second, the sheer pace of recent innovation means there has been too little focus on fitting new offerings into existing environments. Technology generations used to last decades, now it feels like months. Many companies get lost in the fog of technology discussions instead of focusing on end results and engineers don’t necessarily want technologies, they want insight and solutions. Technology is just one resource (an important one) that gives engineers the insight they need to make improvements.
What is needed
The world of big data, predictive analytics, machine learning and cloud computing needs to be turned inside out — from a technology-centric revolutionary approach to a user-focused, problem-solving evolutionary approach.
Engineers can’t just start by ‘grabbing a bunch of sensor data’ — it isn’t a trivial task. It’s also often the start of a longer process that involves cleansing, adding context and performing calculations — a process that needs to leverage the hard-won insights and institutional knowledge of engineers.
Analytics solutions for process manufacturers should fit into existing information system infrastructures and enhance them rather than forcing the enterprise to continually catch up to adapt the latest technology. Process control systems, historical production records, ERP systems and specialised applications and processes are at the heart of your business. Process manufacturers need analytics solutions that can easily fit into these environments through interfaces to their existing systems and enhance their value by extending their functionality and allowing the data they collect and produce to be easily used and shared in new ways. It is better to leverage existing investments than to continuously make new ones.
4. Does the analytics solution require you to move, duplicate or transform your data?
Why it matters
Contextualisation has always been difficult with process data, often requiring manual effort and painstaking work in Microsoft Excel to define relationships between relevant data. Historians have come a long way in terms of trend viewing and investigation, but ‘Export to Excel’ is still every historian’s most important feature for doing the ‘real work’ of data aggregation, context and modelling.
For example, a pharmaceutical engineer might have several hypotheses to explain a bad batch outcome, ranging from an error by a particular operator to a bioreactor maintenance event to a specific raw material variation. The data exists to validate these hypotheses, but it requires bringing together disparate databases, often across multiple data silos, and then creating context to evaluate the data.
This contextualisation process goes by many names — including data wrangling, data harmonisation and data blending — but for many analytics solutions, these tasks still require manual data transformation and duplication. Duplication adds hidden cost to the environment by increasing storage requirements and bandwidth consumption.
What is needed
Users achieve meaningful results when they can focus on data analysis instead of data collection and processing. Analytics solutions for process manufacturers should easily aggregate data from disparate sources and perform the complex calculations needed to synchronise data points that were collected at different intervals. It should also allow engineers to contextualise data without getting IT or other experts involved, without duplicating or transforming the data and without creating additional data lakes.
5. Can the solution help your engineers work as fast as they can think?
Why it matters
Engineers typically look at data for a specific reason. For example, because an alarm went off in a system, or someone has asked a question or they need to generate a report. Traditional analysis tools often require specialised skills or syntax, so these tasks can be difficult and time-consuming — and the tools are typically only mastered by a few people within an organisation.
Beyond the struggles of individual users, few tools are built around collaboration and organisational knowledge capture. When one user cleanses data for a project or creates context and relationships among data sources, that analysis and information is often lost, with no way for other users to discover or leverage it.
What is needed
Data analytics solutions should be flexible enough to support both real-time collaboration and existing workflows. Engineers should be able to interact with tools spontaneously, as quickly as they can create tasks or devise hypotheses. The user interface should allow Google-like searches instead of requiring users to learn a new environment.
Insight gains value when it is shared. For example, one person might know a certain set of process data really well, and they know how to clean and transform that data, while another person might know the ERP system really well, and another might be an expert with maintenance systems. Analytics solutions should make it easy for engineers to distribute their work and provide a central place where employees can work together to leverage each other’s expertise and information.
Presumably, the goal of any analytics solution is to improve outcomes in yields, margins, quality and safety. So, any data analytics solution should be drawing from all of these recent technology advancements to accomplish those outcomes — without your organisation having to enlist expert assistance or know exactly how these underlying technologies work.
The large and growing gap between data and insight in process manufacturing organisations will only start closing when data analytics vendors start putting the process engineer and analyst, by whatever title, at the centre of the picture. These engineers have the expertise, ability and incentive to ask the right questions and take advantage of insights generated by the answers. Analytics solutions need to unlock process engineering knowledge in a way that is convenient for users.
Python is now increasingly being used in the industrial automation and embedded systems world.
Continuous condition monitoring powered by the IIoT is helping many manufacturers in...
In the Fourth Industrial Revolution, self-learning algorithms for autonomous engineering,...