Three phases of implementing best practices for plant benchmarking
Identifying and implementing operational best practices ensures top performance across plant lines, workforces, and sites. True best practices are more than a collection of tribal knowledge, and cobbled together data. Instead, they are built on reliable data that you can compare across plants, for trustworthy extraction and extrapolation of best practices. But in a complex manufacturing ecosystem, these data-driven best practices are not always easy to uncover.
Uncovering Your Best Practices
Plant benchmarking is your best way to uncover best practices that already exist in your facilities, and replicate them across the broader enterprise. Benchmarking enables the comparison and measurement of your productivity, processes, shifts, lines, and other performance-influencing metrics across plant floors, with inherent visibility into what drives superior, repeatable performance. Reliable, apples-to-apples benchmarking depends on normalised data and standardised reporting metrics. But lines and plants are inherently different, resulting in skewed and unreliable benchmarking.
Based on PTC’s and InVMA‘s industry expertise in helping manufacturers like you uncover best practices, we’ve broken the benchmarking process down into three distinct phases: Discovery, Analysis, and Enablement. Each phase answers specific questions and includes distinct goals. Read this blog article to uncover not just today’s best practices, but also enable continuous, scalable benchmarking that can adapt as your plant floors evolve over time.
Phase One: Discovery
In your discovery phase, focus on collecting the data that can best align to your specific goals, and identifying what supplemental data you need to meet those goals.
The key component of the discovery phase are integrated and aggregated plant data. Successful data discovery presents all the data you need, in an accessible format. Data integration and aggregation brings your benchmarking data together in a structured way—laying the foundation for holistic analytics and normalisation.
Common challenges of these phase include disparate pieces of equipment and business systems across various plant sites storing different data and running on unique protocols. These disparities can complicate data integration. If you don’t efficiently connect and sort plant data dashboards, you may end up with silos of non-contextual data. Integrating and aggregating data across plant floors and plant sites can be expensive, time-consuming, and sometimes just impossible.
Questions to Ask During the Discovery Phase
• What am I hoping to analyse and what is the best way to view that data?
• How can I ensure that variances in plant and line equipment ages, levels of automation, and data availability aren’t hindering my data collection?
• Does my data aggregation incorporate traceability, or otherwise include the ability to trace back root cause analysis?
Discovery Phase with the ThingWorx IoT Platform
Through industry-leading data connectivity, ThingWorx integrates and aggregates data from disparate equipment and systems. This includes protocol adapters and aggregators for seamless connection, integration, and aggregation of industrial equipment—regardless of equipment age, automation level, protocol, or system. Data can be integrated directly from sensored products, PLCs, IoT gateways on the factory floor, legacy equipment, and from other business systems, such as CRM and ERP. Once integrated, data is then accessible through aggregated, structured dashboards, and ready for in-depth analysis.
Phase Two: Analysis
Now that your data is integrated, the Analysis phase ensures that data is accurate and actionable. Data integration and aggregation aren’t enough on their own—proven data completeness is necessary for accurate and holistic benchmarking.
The key component of the analysis phase is automated, reliable data connectivity. According to a manufacturing survey by IndustryWeek , 82% of those gathering and processing data through automated dashboards report “complete” or “mostly complete confidence.” Alternatively, only 40% of those using manually assembled data had the same levels of confidence.
Reliable and complete data—achieved through automated, single-view dashboards—is key to trustworthy analysis.
However, even integrated data can be rendered all but useless due to unpredictablegaps caused by communication loss, latent or inaccessible data, and non-standardised formats. If your data can’t be trusted to accurately represent all variables, no plant benchmarking comparisons can be made.
Questions to Ask During the Discovery Phase
• How confident am I in my plant data?
• How automated is my KPI reporting?
• Is my plant data flexible and agile enough to incorporate continuous improvement initiatives?
Analysis Phase with the ThingWorx IoT Platform
Equipment and network designs are unique to every plant floor; the ThingWorx platform was built with the flexibility to address these differences. ThingWorx easily connects throughout any plant floor environment, quickly sources real-time data, and includes multiple redundancy options to ensure communications resiliency. You can always connect, manage, monitor, and control data through one automated, single-view dashboard that maintains reliable communications with critical plant floor components
Phase Three: Enablement
Now that you’ve ensured reliable data integration and communication, you can uncover operational and performance best practices in the Enablement phase.
The key component of this phase are in-depth, customisable analytics. The reliable data you produced in the first two phases now serves as a solid foundation for extracting and extrapolating plant best practices. Analytics are vital to building normalised data and standardized reporting atop that foundation. The right analytics will both reveal relevant best practices and show how to incorporate them for maximum benefit.
However, even with reliable data, determining the best source for the most relevant benchmarking is difficult. For example, if your KPIs are measuring a niche machine, there’s likely too little data for a statistically relevant sample. But if the KPI is too high-level, it isn’t actionable. It’s often difficult to create contextual, analytics-based views of measurable metrics.
Questions to Ask During the Enablement Phase
• How am I standardising data and reports from different plants, lines, and shifts?
• Am I confident that my plant benchmarking is creating apples-to- apples data, in the right context?
• How will these KPIs impact performance across different levels of the enterprise?
Enablement Phase with the ThingWorx IoT Platform
ThingWorx analytics uncovers best practices in your data by digging deep into root causes. Automated KPI reporting removes manual workflows, which saves time and money, and ensures metric consistency.
In-depth data and analytics on a single dashboard gives you full views for monitoring initiatives and adapting, while normalising your data so you can make apples-to-apples comparisons. You gain reliable insights to confidently understand why top performers are doing better than others—beyond just speculation—for a data-driven, highly performant plant ecosystem.
Proven Best Practices for Improved Performance
Designing your plant floor around proven best practices reduces costs, improves productivity, and provides a competitive advantage. ThingWorx integrates, aggregates, and normalises your data for relevant, reliable benchmarking all in one platform. You are able to reduce the time and effort required to report on KPIs, while identifying best practices and improving performance within and across your plants.
The Future of Multi-Plant Benchmarking for Improved Performance
Review Key Survey Findings About Multi-Plant Benchmarking
Manufacturers have a lot to gain from implementing plant benchmarking initiatives—including enhanced operational performance, greater data visibility, more data-based best practices enablement, and more.
In The Future of Multi-Plant Benchmarking for Improved Performance research paper, IndustryWeek breaks down key findings from their 2018–2019 survey of manufacturers. The report explores:
- Challenges, best practices, and other results driven by multi-plant benchmarking in modern manufacturing
- The role of IoT in confident, successful benchmarking data collection
- IoT investment trends, and how different manufacturers are choosing their investment strategies