Using a "Digital Twin" to Optimize Your Company Data

What is your data communicating to you? Does analyzing your data feel like trying to drink from a firehose? What is the TRUE value of your current data? In this article, learn about how adopting a “Digital Twin” can help your company optimize the way you use your data.

Move from drowning data to imagining the future of your products

Companies trying to find a competitive advantage are having to work harder than ever to keep up with their large datasets, and Research and Development and Product Engineering have never been more important in ensuring that a new product meets market demands and manufacturing realities.

A technique used by many top quartile companies is to simulate the manufacturing systems in their ERP system, prior to investing in space, equipment or materials. Sometimes called creating a “Digital Twin”, these simulated operations allow forward thinking companies to learn more quickly, and create evolved production systems that are more efficient and flexible.

The power of machine learning

Machine learning, tied to Big Data and Internet of Things (IOT) sensors powers many of these new approaches. Companies monitor real time manufacturing, sales, and supply chains of existing products, and use this information to create new products, as well as variants of existing products. Factors including machine speed, seasonal weather, and quality checks can be regrouped to provide meaningful information for current, and future products, instead of wasted data, or worse yet, noise. These advantages have been known about for some time, and are in use in leading companies in all industries.

Top performing organizations take this a step further - they use this abundance of data, and their ERP and other manufacturing systems to simulate production with enough accuracy to drive out mistakes and waste, and create optimized manufacturing operations, prior to embarking on large investments.

Mapping optimized operational conditions

Simulated operations use advanced analytics to map optimized operational conditions. These optimized conditions can be played out over one day, one month, one year, or over whatever time horizon is best. Supply chain or manufacturing bottlenecks can be identified proactively, and addressed quickly. Equipment breakdowns can be simulated to learn the effects, and with this information, predictive maintenance can be included, reducing glitches on the production floor. All of this can be done prior to investments in labor, space or equipment, reducing risks measurably.

Costs and the task ahead

All good things come with a start-up cost. When considering cost, isn’t it better to spend thousands now to save millions in the long term? Properly understanding the start-up costs of an improvement is essential for long-term, sustained growth. In order to set the framework for a robust digital twin approach to business growth, your organization’s processes and procedure must be accurately mapped and documented. The power of a digital twin is only realized when the source information is robust and accurate.

Implementing a digital twin is no small task. It requires each transaction from order to cash and procure to pay to be identified, investigated, and validated. Mapping these transactions in a process map is the first critical step. Once processes are mapped, they can then be quantified in terms of cost and value. Each step is now clearly understood, and each actor (machine or human) that is responsible is accounted for. The output of these transactions is always data, and the value of that data is more clearly understood when mapped at the organizational level. Do you know if the Open Orders report generated by Customer Service is actually of value to Production Planning? Is there another report that the planners use instead? A digital twin can help you quickly answer these questions in an efficient manner.

Once your business processes are digitized and quantified, you have a digital twin. This can be used to simulate change. One simple change can be proposed to the organization and all downstream and upstream impacts can be identified. Eliminating the Open Orders report may save Customer Service 10 hours per week, but it may also impact how Purchasing evaluates current raw material demand. A digital twin can quickly identify this unforeseen consequence and ensure that valuable data for one business unit is not deleted for the sake of optimizing another. The true power of running simulations, such as the simple one described here, is to not only identify opportunities for improvement but to also identify potential consequences.

When adding a new product line, your digital twin is essential to understanding ALL costs and consequences associated with the change. Can the new product be manufactured with the current configuration? Does the new process impact another process in an established product line? What are the potential operational and transactional costs of implementing or changing a product line? Do these costs align with those used to calculate the ROI? Once again, simulating these changes first can incur a small cost, but can also mitigate the risk of incurring much larger costs due to unforeseen consequences.

Companies that have implemented this approach positively have seen 5% or greater growth in productivity, lower costs, and measurably better quality. These insights have unlocked capacity in factories, eliminating the demand for more space or equipment.

Open Source Integrators’ experience in this is significant, dating back decades, using ERP and other associated tools to create intelligent simulations that drove key construction, utility, and manufacturing decisions. OSI’s decision support systems (DSS) have helped business and policy leaders avoid problems, and make better decisions, with more success.