Auto-Generated Dynamic Recipes + Productivity KPI Model: Fall Product(ivity) Release

This summer, we have been working hard to incorporate new features requested by our customers and break some new ground in technology for manufacturing productivity. We are rolling out two new features of note that I describe in this post – Golden Run Recipes and Real-Time KPI measurements.
Dynamic manufacturing

Table of Contents

This summer, we have been working hard to incorporate new features requested by our customers and break some new ground in technology for manufacturing productivity. We are rolling out two new features of note that I describe in this post – Dynamic Manufacturing Recipes and Productivity KPI Models. Both are unique to Sight Machine and both help a manufacturing operation shift their analytics and operational management to real-time. Both can deliver real gains in productivity and OEE, even in a matter of weeks.

Auto-Updating Dynamic Manufacturing Recipes to Maximize Productivity

Dynamic Manufacturing Recipes

Every process engineer wants to be able to understand what settings on a line deliver the maximum productivity – the “Golden Run” recipe for optimal operating instructions. The reality is that in most manufacturing operations there are a variety of variables that can impact all the components of productivity: quality, performance, availability, and cost. For example, a company making glass bottles might find that the recipe for a Dynamic Recipe of line settings is different in conditions of high humidity or low humidity. Similarly, a copper smelter may have a different Dynamic Recipe depending on the chemical properties of different types of copper ore. Traditionally, process engineers update Golden Run operating instructions yearly because calculating Golden Runs is time-consuming and involves a lot of manual work. Often operators distrust Golden Run recipes and operating instructions because these instructions do not consider wide variances in operating conditions – temperature, humidity, differences in material.

Sight Machine built Recipes to address this problem and make it easy for process engineers to provide accurate Dynamic Manufacturing Recipe operating instructions considering multiple scenarios and context. Process engineers and production teams can organize their Recipes into Cookbooks. Recipes and Cookbooks empower process engineers to quickly identify and capture Golden Runs based on any desired outcome and given any set of conditions. As well, Sight Machine automatically updates existing Recipes by analyzing real-time data to reflect newly identified Golden Runs.

The Sight Machine Dynamic Manufacturing Recipes and Cookbooks feature offers the following capabilities:

  • Simple menu-driven path to identify the best operating instructions for the desired outcome including production volume, quality, or cost
  • Automatically generated operating instructions for multiple Recipes based on the process engineers formulas
  • Grouping of related Recipes into Cookbooks to enable process engineers and operators to easily maintain and update Dynamic Recipe instructions for multiple products
  • Automatic updates in real-time of Recipes based on newly identified Dynamic Recipes

What This Means for Manufacturers

For manufacturers, Recipes and Cookbooks offers two core improvements. First, providing accurate and data-driven recipes that are updated in real-time ensures that every operator is running their machine and every shift supervisor is running their line in the most productive way. Effectively, Recipes and Cookbooks are capturing all the tribal knowledge stored in the heads of operators, validating which of that tribal knowledge is the most beneficial, and compiling the best tribal knowledge into Dynamic Manufacturing Recipes. Second, by making it far easier to analyze data and generate operating instructions that take into account detailed real-time conditions, Sight Machine’s Recipes and Cookbooks give manufacturing teams the capability to optimize and account for a wide variety of factory contexts. Ultimately, this can help manufacturers improve productivity and OEE without making any changes in equipment, plant, or requiring major training.

The Benefits of Recipes and Cookbooks

Capturing this data and transforming it into knowledge provides a sustained advantage to manufacturing companies facing increasing levels of retirement and subsequent challenges on onboarding newer workers. Dynamic Manufacturing Recipes and Cookbooks also can bridge the trust gap between operators and process engineers.

  • Plant teams using Sight Machine can more quickly onboard and train new operators by using the Recipes as training tools and stores of “tribal knowledge.”
  • Operators can trust and more easily follow provided operating instructions because they are based on detailed data analysis and updated in real-time to reflect the most current conditions.
  • Shift leaders and plant managers can reduce the impact of change orders by providing easy to follow operating instructions for all products in a single well-maintained repository.
  • Process engineers can make “Continuous Improvement” proactive by identifying the best operating instructions and pre-empting inefficient tribal knowledge with data-driven insights.
  • For CEOs, COOs and CFOs, Dynamic Manufacturing Recipes and Cookbooks can deliver sustained improvements in OEE and productivity merely by ensuring that every operator and every plant manager has the best guidance and instructions on how to operate their line.

New KPI Model, Widget and Dashboards

dynamic manufacturing

For manufacturing companies to date, measuring KPIs in real-time and at the level of actual machines has been challenging. We wanted to take these measurements deeper to provide better insights to the workforce on the shop floor by showing them, in real-time, whether they are on plan and illuminate ways to boost productivity, quickly. We also wanted to give management teams the ability to compare KPI performance on any selected parameters easily. In our view, real-time and comprehensive measurements of productivity and KPIs are the clearest path to continuous improvement because they empower incremental gains in productivity at every level of the manufacturing process.

The Sight Machine KPI Model offers the following capabilities:

  1. Provides real-time updated KPIs which allow teams to track production numbers up to the minute
  2. Ties KPIs to underlying data models to make it easier to answer questions such as “How does OEE differ from one supplier or plant to another?” and “How does humidity affect quality”?
  3. Enables objective comparisons of productivity across varied and dissimilar data sources (e.g., machines of different types, age, capacity or throughput ratings)
  4. Empowers processes for network-level and enterprise-wide continuous improvement by making it easy to identify, visualize, and track variances from theoretical max

What This Means for Manufacturers

With the KPI Model, manufacturing teams can quickly add KPI indicators to real-time dashboards or visualizations already in use. The KPI Model allows for expanded use of KPI metrics across an entire plant workforce and for plant managers, executives, and others who may not be on-site. (In other words, real-time remote KPI monitoring!) Machine operators use a nearby monitor to view a simple widget showing performance on KPIs. Continuous improvement, process, quality engineers, or manufacturing heads can pull up dashboards with embedded KPIs on their laptops or mobile devices.

For plants with assets of varying ages, the KPIs can be customized to reflect different asset performance expectations based on theoretical maximums and taking into account environmental conditions. For example, an older generation extrusion or stamping asset might have a lower theoretical max than newer assets performing the same tasks. Those older extrusion machines may be more susceptible to fluctuations in ambient temperature. KPIs can be adjusted in the data model to reflect these conditions related to theoretical max – making those KPIs dynamic and automatically adjusted to reflect reality.

This granular treatment based on theoretical max enables apples-to-apples comparisons and KPI measurement and benchmarking that aligns with the age, state, and type of machines on the shop floor. Part of why we built this feature is that our customers wanted to find opportunities to improve OEE and their plants’ performance. Once a manufacturing team knows each asset’s theoretical max and can visualize and track performance, it is much easier to find areas for improvement.

Benefits of the Sight Machine KPI Model

From these capabilities, manufacturing companies can construct accurate real-time OEE tracking.

To simplify all KPI reporting and make comparisons more useful, we made it easy to configure the reports’ timeframes. It is push-button simple to select a timeframe and compare KPI performance.

  • Quick configuration of KPI reporting to match your production and reporting periods. Set to “start of week,” “start of the quarter,” and “start of the year” with a few clicks. Quick configuration enables manufacturers to track KPIs in real-time according to their specific production schedule and to track different production schedules for different units with a global view.
  • Add calendar shortcuts such as “This Quarter”, “Last Quarter” and “YTD” for KPI reporting and benchmarking on defined and rolling intervals. This allows teams to toggle between KPI views quickly and to analyze trends in the short-term and long-term.
  • A single-value real-time KPI widget to keep operators on the shop floor informed and draw their attention to the largest opportunities for improvement. The single value KPI widget gives operators a clear view of the information they need most and helps them track performance towards KPI goals.

Dashboard Bulk Edit

Productivity KPI Model

Many manufacturers waste a lot of time trying to reconfigure dashboards or other data outputs to perform accurate comparisons. At worst, it means pulling data from historians or SQL databases and cleaning the data. Even when the data is in visualization platforms, each dashboard must be reconfigured manually to enable comparisons. We built the new Dashboard Bulk Edit to eliminate this problem for our customers.

Benefits of Dashboard Bulk Edit

With Dashboard Bulk Edit Sight Machine customers can:

  • Edit parameters on some or all dashboards in a single step
  • Quickly dig into assets or specific time ranges within bulk edit to make quick investigations possible (during shift changes, for example)

Bulk edit is also useful for machine OEMs.

More OEMs are now actively adding service and remote management capabilities to their product offerings. With the Dashboard Bulk Edit feature, OEMs gain powerful capabilities to add new fields and parameters to deployed machines or reconfigure existing dashboards quickly and easily to reflect customers’ needs. Bulk Dashboard Edit also will allow manufacturers that deploy machines to end-users – for example, box making machines – to more easily investigate what is happening and perform both predicate and root-cause analysis, at scale.

More Cool New Features Coming Up!

This release moves Sight Machine customers closer to the vision of having digital transformation accessible to every employee by giving them accessible and easy to understand KPIs to guide their performance. It also allows management teams to finally compare on a level playing field asset and team performance across shifts, lines, plants, or the entire global networks of facilities.

Everything we do is designed to make manufacturing teams’ lives easier and better, from the shop floor to plant managers to regional and global leadership teams. Making manufacturing stronger, sustainable, and resilient means giving the people that make what the world needs the very best technology to do their jobs and improve their productivity. Thanks for reading.

Andrew Home

Andrew Home

Andrew Home is a data scientist and product manager with a passion for understanding the relationships between data and manufacturing processes. Andrew has served as a data science fellow with both Cap Gemini and Galvanize Inc. He holds a BA from Southern Methodist University.

Curious about how we can help? 
Schedule a chat about your data and transformation needs.