When the idea of PLM systems first time was introduced, one of the main drivers was to develop an infrastructure for collaboration and for sharing information among the company users during product design, engineering, manufacturing, and sales. Such an idea of a single source of truth, breaking siloed systems was (still is) very popular among companies looking how to develop their PLM business initiatives.
Manufacturing companies have been struggling with data silos for years. Different departments within the company would store their data in isolated systems, making it difficult to get a comprehensive view of the entire manufacturing process. This made it difficult to make strategic decisions and left manufacturers at a disadvantage compared to their competitors. However, recent advances in data management and cloud technology have made it possible to bring new and innovative methods to break data silos and create information streams that provide a real-time view of the entire engineering, manufacturing, and maintenance process. In my blog today, I will discuss how using web services and other modern data management technologies, manufacturers can finally get the insights they need to make smart decisions and stay ahead of the competition.
How complex is a single source of truth?
A single source of truth is a powerful concept. Although it comes these days with a grain of salt. The biggest challenge is that companies are struggling to walk a complex path from the PLM idea to the reality of the implementation. PLM systems that are well established for large companies are facing struggles when using the same implementation strategies for SMEs. At the same time, large and complex enterprises are struggling to create cohesive data management strategies to apply to multiple enterprise systems – PLM, ERP, CRM, MES, etc. Enterprise companies have to choose between a single vendor blackhole and managing multiple systems with sometimes no very open data interfaces and complex integrations. This is a reality of PLM implementations in the 2020s…
Breaking Silos? What Silos?
Manufacturing companies have been struggling with data silos for years. Traditionally information was stored in each individual system – designers used their desktop-based CAD/CAE systems. PDM/PLM was very focusing on engineering data management. Manufacturing planning and procurement were gravitating towards advanced usage of MRP/ERP systems. Sales are living in their own silo and maintenance and support was another siloed data space.
In my view, there is a significant misunderstanding about what means to break siloes in an organization. Before jumping into the discussion, let’s first define what silo are we talking about. In my view, every organization (regardless of size) has its functional siloes. Even a small organization with 5-15 people will have functional silos. From a process standpoint, the engineering process is separate from a sales team, marketing is separate from the supply chain, etc. You don’t really want to have all these people together because each of the smallest departments is operating with its own goals and process. However, the biggest problem is happening when data silos occur and these organizations or individuals don’t have seamless access to the information they need.
Information Flows and Data Handover Problem
The biggest problem in many manufacturing organizations is the de-facto existence of data silos. What happens very often is that each organization (or individual in a small organization) is organizing information in the way they prefer. In smaller organizations, you can CAD files, spreadsheets, and sometimes finance applications. In larger organizations, you can see PDM/PLM systems, ERP, and financial tools. These applications are obviously performing the functional roles for each of these groups of users, but when it comes to the data, the organization is drawn into a data mess and siloed data spaces.
At that point, manufacturing organizations are starting to look at how to solve the problem by either introducing a “single system” or inventing ways for a company to integrate these information silos together. The problem of data handover is becoming urgent because the organization is losing efficiency, working on discounted and often outdated data files (systems), and eventually not performing to its best. As a result, mistakes in design, mistakes with orders, wrong data for certification, and many other problems are coming up very quickly.
From Traditional PLM Systems to Online Web Services
A traditional solution with data siloes I described above is to have a single system that organizes the data and by doing so, kills the silos. As much as it sounds like a good approach, it is not that simple to organize. Even a very small organization, might be in disagreement about functional needs for that “single system”. There are multiple functional needs. When coming to a large organization the situation is very complex. Until now, the industry provided a simple answer to that problem – multiple systems + integrations. The approach can work and a solution will go down to the path of figuring out how to extract data and transfer it between multiple systems using files.
What can be a better approach in the future? SaaS, web services, and RESTful APIs are coming. Manufacturing companies will be adopting web services and use standard REST API to communicate between multiple systems. The same approach can work for both small and large companies because REST APIs provide a ubiquitous and seamless way to access data. At the same time, SaaS applications will be scaling from multi-tenant SaaS platforms to complex enterprise SaaS solutions when SaaS platforms will be seamlessly available via dedicated solutions available via PaaS providers (AWS, Azure, GCP, and others).
What is my conclusion?
The problem of data silo and data integration is super critical for product lifecycle management. But silos are not about making everyone work together. Functional silos will stay. The problem of gathering data, organizing data sharing, and getting people to access relevant data at the right time is super critical for many organizations. Modern business processes are quire exchange of data and better organization of data assets, which by itself brings the question of data governance, organizations of data sets, and customer data on the top of requirements for many organizations. The old recipe of having a single pace for all data (e.g data warehouse) is good for some use cases, but completely useless when companies are looking at how to integrate software systems in real-time. Web service and SaaS platforms are growing their adoption and more and more business leaders are choosing these modern technologies for digital transformation programs in these companies.
Just my thoughts…
Disclaimer: I’m co-founder and CEO of OpenBOM developing a digital cloud-native PDM and PLM platform that manages product data and connects manufacturers, construction companies, and their supply chain networks. My opinion can be unintentionally biased.