For many manufacturers, the Factory of the Future is increasingly a sustainable production facility, using digital tools and data insights to do more with less – operations are guided by data and enabled by automation, achieving the same outputs but using lower inputs. And doing it at scale.
Developing a Sustainable Factory of the Future means adding new digital technology to traditional methods, adapting processes to operate sustainably, and adopting a culture of experimentation and innovation to investigate new approaches that constantly improve ratings. Here are two real-world examples:
- Lenovo saves over 2,696 MWh of electricity annually, that translates to an emission reduction of 2,000 tons of carbon dioxide
- Automotive supplier Gestamp reduces its energy consumption by 15percent by harnessing Big Data 2
How did they make these gains? The answer is they both adopted advanced digital technologies and used data insights to better manage a key resource (power) during manufacturing.
Start With Cost Reduction, Get Sustainability as a Side Benefit
For many manufacturing facilities, significant automation and digitalisation initiatives are already underway to move from proofs of concept to scaled out industrialisation. Whilst these initiatives may not have been initiated by Sustainable Production objectives (cost reduction and yield improvement, are often the drivers), sustainability targets can also benefit.
Programmes that seek to be more efficient with consumption of electric power in a factory are explicitly sustainable as they reduce inputs (power consumption) whilst maintaining output. The driver may have been cost reduction, but the outcome impacts two targets.
How might power usage reductions be achieved? There are several options including better monitoring of idle time, reducing unplanned outages, moving to smart illumination, to name a few. What they all have in common is smart use of data.
Reducing Power Usage is Big, but What Other Challenges Can Digitalisation Impact?
Being smart with data doesn’t just help with power consumption. The following key manufacturing metrics can also be positively impacted:
- Cycle Time – reduce cycle time by improved cleaning and maintenance interventions.
- Lead Time – reduce overall lead time by smarter supply chain operations & tighter integration.
- Inventory Holds – reduce inventory levels as supply chain visibility improves.
- Right First Time – reduce rework, quarantine and expiring stock with improved planning models.
A big sustainability win for industrial producers is avoiding rework or remediation efforts. These are pure costs to the business with no revenue increment for the increased effort. They are also expensive in resource usage (e.g. materials, power, and time). By having accurate data on current and historic operations, any variance from operating norms can be detected early (or even anticipated) and light-touch interventions initiated.
These remediations will often be quicker, easier, less toxic and less power-intensive than if the error were discovered later. Errors that go undetected can have a troubling impact on both yield and quality, whereas using advanced data models can help return operations quickly to nominal operating conditions, minimising impact on yield and quality.
By gathering granular measurements about manufacturing operations and referencing historic norms and trends, a rich base of data is created. Computer models then scan through these vast amounts of data and pick up on small variations, identifying trends and opportunities for improvement. Computer models can analyse more information than humans, spot trends faster and predict future outturns with more accuracy. In fact, the scale of data available in most manufacturing facilities often overwhelms our ability to scan and act, without the aid of digital tools and methods.
Being data led can provide immediate benefits in detection of abnormal operation and remediation. Additionally, every subsequent cycle improves the predictive strength of the computer model and thus the recommendations improve over time, yielding both commercial and sustainability improvements.
Incremental Changes Add Up
Big wins within efficiently run facilities may have already been realised; however, smaller scale improvements can still deliver worthwhile outcomes, and help on the path to sustainable production. Some of these may be quite easy to find. For example, taking a data-led approach in a unit operation (such as a drying step) allows a narrow focus for a process improvement exercise.
Projects like this are often easier to get commissioned and to keep track of and demonstrate progress, without diverting large teams from their prime task of managing daily output.
Staff experience using data models is transferrable and amplifies across the organisation. For every data model and data visualisation shared, more team members gain experience of data-led approaches and get comfortable using predictive analytics to positively impact day-to-day operations.
Not Just for Efficient Operation, Sustainability is a Huge Focus for Governments
With the recent COP26 meetings, Scottish and UK governments have demonstrated that they are very active in pursuing both citizen policies and industrial strategies that focus attention on climate impacts and drive change to more sustainable models. As the years pass, support and encouragement from government likely turns increasingly to monitoring and enforcement as target dates for delivery of a net zero objective approach.
Getting ahead of the curve will help manufacturers do their part in achieving net zero, but also positions them to be more efficient and data-savvy operators, which could increase competitiveness. Bear in mind that other national governments will be providing support for digitalisation to their manufacturers, so delaying adoption of improved data strategies could be competitively damaging.
Benefits Are Not Just for Large Producers
Whilst larger operators may already have data skills within their organisation and experience of model building and visualisation, the benefits do not accrue only to big organisations. Smaller firms can benefit greatly by being data-led and can gain significantly in their sustainable production goals by starting small and learning fast.
Whilst small producers may lack the resources of larger organisations, they have the advantage of simpler decision making chains and can quickly promote process improvements. Similarly they can ditch unproductive ideas and quickly pivot to the next idea. So data-led sustainability programmes may succeed faster in smaller, more agile teams.
Some preliminary steps that will enable smarter use of data in manufacturing:
- Automated collection – migrating from manual collection and transcription to automated collection and recording
- Historian archive – maintaining historic data records for extended terms to compare different periods and note changes to operations, instrumentation, scheduling, etc.
- Data quality detection – when data quality changes, e.g., if a gauge or meter is faulty or if a feed from a third party introduces errors, being able to identify and rectify quickly
- Skills development – developing skills across the workforce to become more comfortable with data, data analysis and predictive models
Becoming data-savvy for sustainable production can be a daunting prospect but there is lots of help available. Organisations such as DataLab & National Innovation Centre for Data (NICD) have skilled teams ready to help organisations to better manage data through skills development, collaboration, and advice.
Training and education from leading providers such as National Manufacturing Institute of Scotland (NMIS) and organisations such as Royal Statistical Society, provide a wide range of courses to cover core skill enhancement and application in industrial settings.
There are also many commercial organisations offering training in data related areas, such as coding in R or Python, managing large cloud-based data sets and visualising data in Tableau or PowerBI. Plus, a range of consultancies are available to provide skill augmentation, to take on development efforts and help grow your capabilities with knowledge transfer. For an informal chat, please contact us.
Whilst tools such as Tableau and PowerBI can be great at extracting insight from data (and there are many, many more available), it’s worth remembering that tools can only work with the data available to them, and if the data quality is poor, the tool can do little to improve that.
That’s why many data analytics projects become “data collection and data quality projects”, before yielding their analytics outputs. The journey to net zero may involve data quality improvement as a precursor to data analysis.
1.Manufacturing Global, Lean Manufacturing —
2.Saving Energy With Big Data