UncategorizedNo Comments

default thumbnail

Data is quickly becoming a core element of every business these days. In my opinion, manufacturing is clearly a big underdog when it comes to the question of data management. The data in manufacturing companies is heavily siloed between multiple systems, departments, companies, suppliers, contractors.. you name it. At the same time the data, the way it is gathered, stored, and consumed during business processes can determine the success or failure of the business today and tomorrow.

The question I want to raise today is about data quality, why is it important, and how modern SaaS data management and PLM systems can make a difference in manufacturing data systems moving forward. There is a clear opportunity to get insight into various aspects of manufacturing processes by bringing data quality to the next level by simply organizing one of the most important assets manufacturing companies already have – data.

Manufacturers are inundated with data from multiple sources and databases. While core business systems such as engineering (CAD/PDM), enterprise (ERP) and manufacturing execution system (MES) hold some data, the amount of data outside of these systems is such huge that raises the question of data quality overall. Recently, I observe a growing interest in manufacturing companies to the concepts of Digital Twin and Digital Thread. These concepts are interesting, but unfortunately often rely on old and unreliable PLM infrastructure developed a few decades ago. In most cases, digital twin and digital thread are great marketing concepts selling the same PLM technologies we’ve seen for the last decades.

So, what are the root causes of data quality problems and how to improve it? In my view, there are three main sources of problems:

1- Data incompleteness in silos

With a large number of data silos, systems often are lack data. Especially if the data is originated in multiple places. To bring this information between systems is hard and companies often rely on partial data representations. For many years, meaningful (intelligent) Part Number is one of these elements that companies are using to compensate for data incompleteness. Various elements of Part Number are used by a company to identify, type of the component, supplier, version, and other data semantics.

2- Manual data re-entry

Integrations are hard and often companies need to re-enter data between systems. A typical BOM report from CAD system used by procurement to re-enter data into ERP system. This is only one example, but each data re-entry is a source of potential mistakes and errors.

3- ETL and data synchronizations

ETL which stands for Extract, Transform, Load is a way to move data between systems in an automated way. While automation is a good part, mapping data between different systems is a source of data mistakes, wrong transfers, and missed transactions.

So, what can be done about it? For years, siloed systems of data management in manufacturing companies remained almost the same, and Excel massively filled the gaps used by many chief excel officers. What tools can be available for modern digital data plumbers and why it can be different from everything we’ve seen before?

The fundamental change between yesterday and tomorrow is in the way data is treated by manufacturing companies won’t happen overnight. The foundation of the change is the data management capabilities to hold the core of manufacturing – product data. You can call it product structure, bill of materials, system engineering. I will leave it to the marketing of these companies to develop the solution. However, the models developed by these systems should conform to these important requirements.

  • Holistic and global data model representing product information. The information should be shareable globally and not locked in a single OEM or Tier n company
  • Scale from very simple models bottom-up to complex products and manufacturing systems. While some product standards made huge progress over the last decades, almost all of them ignored the need required by global online systems capable to operate in real-time.
  • Data share in real-time downstream to provide a single point of data access to other services and business applications.

Combined together, these systems will provide a way to grow a new foundation of data management for manufacturing companies and will eliminate old fashion siloed data management.

SaaS PLM systems are good candidates for the role of a modern data management foundation for manufacturing companies. Built using web technologies, polyglot data management principles, and capability to scale as a global system. A core element of the SaaS PLM – product information model of these systems can provide services to downstream applications and business systems.

During the last four years, I’ve been building openbom.com (disclaimer, I’m CEO and co-founder) – a novel data management system using data management principles described above and learned many lessons of how to scale data models and systems to be used by individual engineers and manufacturing companies globally. My BHAG (big hairy audacious goal) is to build an open system capable to connect manufacturing companies in a network that can be operated using data intelligence mechanisms built using product information.  You can read about my ideas for future PLM circa 2020 here.

One of the advantages of modern SaaS PLM technologies is to be connected using web service technologies to consume data. In such a way old fashion PLM system with the business of data locking will become a thing in the past. Modern PLM SaaS systems will have a much easier mechanism to integrate and connect with each other.

What is my conclusion?

It is time to rethink the core foundation of data management for manufacturing companies. Product information systems combined with modern multi-tenant cloud system architecture can become a core element of new manufacturing information providing services downstream and to all other business applications. Eliminating data silos combined with the ability to share data in real-time downstream will significantly improve data quality and create a foundation for new business processes in manufacturing. Just my thoughts…

Best, Oleg

Disclaimer: I’m co-founder and CEO of OpenBOM developing cloud based bill of materials and inventory management tool for manufacturing companies, hardware startups, and supply chain. My opinion can be unintentionally biased.

Share

The post How SaaS PLM can improve manufacturing data quality and create a distributed single source of truth appeared first on Beyond PLM (Product Lifecycle Management) Blog.

Be the first to post a comment.

Add a comment