Is siloed BOM a thing from the past? My old article PLM, Bill of Materials, and Silo Syndrome raised a healthy number of great comments and discussions on LinkedIn, so I decided to come back to this topic and discuss BOMs, organizational silos, and data management again.
A dread silo syndrome is not something new in manufacturing companies of any size. The reality is that you need only two people to have silos. You can see it sometimes in small engineering and manufacturing teams when design engineers put a barrier between design information they create and own and the rest of the organization. Production planners and procurement are dependent on the design data and bad communication creates painful processes of data exports, reporting, data consolidation, and synchronization. Both sides have a hard time to agree on how to transmit, share, and consolidate information.
Similar problems can happen on a bigger scale in large companies even if more technology and products will be involved. Siloed design teams, PDM, PLM applications, siloed PLM and ERP applications, and many other systems.
In the modern world, when manufacturing companies are moving towards service business models and companies are looking for an advanced business loop to bring customer information back to manufacturing companies, to monitor what customers do, the problem of silos is expanding and includes product information coming from customers. You need to know exactly what includes the product for each specific customer, what maintenance parts were replaced, and what software and hardware upgrades were made.
With the risk to dive into terminological debates about what is Bill of Materials, I found that for many small and medium-sized companies as well as manufacturing enterprises, a BOM or xBOM or just Bill of Materials is a sufficient abstraction level to describe product information, which includes data about the product and all related pieces of data. So, please accept my apology if you prefer to use words like “model” or “data” in this context. In my view, the Bill of Materials provides a better semantic level, which includes product structure information as a core foundation of the data abstraction model. As such you can see xBOM models including design, engineering, production, cost, maintenance, and many others.
The reality of many organizations is to have data distributed between users, files, databases as well as organizations, and different systems. How to connect them and consolidate these silos? This is a question many companies are struggling with today. Companies are spending tons of resources and efforts on aligning and validating information, limiting unnecessarily duplications of data or re-entries of information. Each of these duplications or re-entries are impacting data quality and making company processes inefficient. Here are my 3 steps on how you can improve data quality and visibility of BOMs in a siloed environment.
1- Data and Information Model
There is an abundance of data management tools and technologies these days. The time when database technology list was limited to a few names with a high price tag is over. You can think about databases as a tool to store data in different forms. There is no single best database. It is like a tool – you need to have the right set of tools. It is called polyglot persistence and it allows you the right technology when managing a specific information model. Unlike a single database system, polyglot persistence became a foundation of large scale global systems capable of storing huge amounts of data. To connect and consolidate siloed BOM requires to build an information model capable of creating a data representation to reflect the needed information. Each company can be slightly different, but an information model should be capable of scale to different sets of information – coming from multiple places. Once it is done, define what data management system or systems are needed to manage it.
2- BOM Control And Sharing Technologies
Configuration management principles is a good foundation to start thinking about elements to process and to control all these pieces of information. A good start is to think about object identifications, attributes, relationships, who can control access, and how data can be shared. A complex environment might require multiple levels of administrations and control. Actually, data control and ownership are the main sources of why companies or people are getting into silos. With no distributed control, the fight over data ownership ends by separating data and organizing silos. Sharing technologies are very popular and can ease the way data is accessed online. There should not be any reason why data cannot be share- the technologies of multi-tenant data management systems are making it easy to organize.
3- Data Integration Technologies
The third element in the puzzle is data integration technologies. The reality of companies today is multiple systems and technologies. Companies invested money and time to build them and organize the processes relying on these systems. In some situations, it is just a matter of time until systems will be replaced. In some other situations, companies will live with different systems and integration tech allowing to extract, transform, and connect to the data is the only way to consolidate the information.
What is my conclusion?
Data silos is a complex problem in modern manufacturing. What was the problem a decade ago as a siloed of data in departments of a single company, now is fast becoming a problem of siloes of data in work between multiple companies – contractors, suppliers and moving even more into silos of data between manufacturing companies, customers using these products and organizations performing maintenance and services. To solve this problem, you need a combination of organizational governance and technologies acting together. The first is needed to organize the process and the second is a set of new products and technologies to solve data management problems. Multi-tenant SaaS PLM and other data management systems can become a foundation to solve the problem of siloed BOM because they provide a model as well as technologies together. Just my thoughts…
Disclaimer: I’m co-founder and CEO of OpenBOM developing a digital network platform that manages product data and connects manufacturers and their supply chain networks.