QR_ have spent well over a million hours authoring, maintaining and conducting rapid and accurate BoM validations for over 100 automotive clients, looking beyond the system to optimise human and managerial processes to further enhance release and order.
Compiled with input from QR_ PDM professionals Daniel Garratt, Evgenios Efthymiou, Dave Slawson, Andrew Houghton and William Smith, this is the second in our three-part series for ‘QR_ essentials’ and will explore the factors informing and influencing the inevitable compromise between speed and intelligence in BoM design and functionality.
In the emerging world of quantum computing the above question might soon become irrelevant, but until you run a quantum BoM you will very likely be called upon to choose between speed and intelligence for your BoM system. There are two main scenarios where you could be called on to make this choice: when choosing which BoM system to use or when choosing when to implement a new feature in a BoM system under development.
Spoiler alert: the answer in both scenarios is you need to balance between speed and intelligence! But how do you go about implementing that balance?
What does it mean to be fast and how important is it?
A simple way of thinking about it in the world of BoMs is to ask “how fast can you retrieve the information from where it is stored, manipulate it, and send it back?” Looking at it from the everyday user perspective responsiveness also plays a role; i.e. “how long will it take before something happens after I take action”.
In a study we carried out and highlighted in our article ’What makes a good BoM system?’, system speed was not found to be the number 1 priority. Indeed, we found that a “fast and responsive” system was the third most important criterion, after having “one master BoM” and “robust and reliable performance”.
What about intelligence then?
Intelligence for a BoM system looks at all the capabilities that allow your system to operate efficiently and effectively. You can think of it as how well your system works in conjunction with your business processes, from guiding the user through the restrictions to generating value.
Let us consider an example: the creation of a new part record. A rudimentary system would let anything through with no restrictions or checks, relying on human checks and knowledge of the business process to achieve the correct result. An intelligent system would apply those checks, potentially even suggesting the right format, or going as far as to pull some useful information to help inform decisions. And here lies the trade-off, as all these will cost you in terms of speed and system responsiveness.
Looking at our ’What makes a good BoM system?’ study, intelligence was never cited as a distinct criterion. Instead, it is recognised as a contributing element across multiple base criteria, ranging from achieving “one master BoM” to having an “intuitive and complete UI/UX”.
Is it possible for a system to be both fast and clever?
Like most things in life, cost is often the main factor that drives a wedge between speed and intelligence. Think back to the scenarios mentioned at the start. If you are choosing a new system, it is very likely that your speed is going to depend on hardware, with faster hardware being more expensive. In the other scenario, buying plug-ins or developing new complex features for your BoM system will cost money. Even if they would potentially save you from costly errors down the line, simply running them requires some daily processing power. Suddenly, your BoM system is now crawling and no-one wants to work in it.
One school of thought would argue that whatever the complexity, validations and workflows that are required to run should be allowed to run and we should work around the time this takes. After all, BoM work is not the only thing employees need to worry about.
Here lies a trap we have seen time and time again. If your system does not deliver immediate access to its data, you can be sure your user will still get it, often through an offline and probably out-dated record! Based on our experience the biggest impact a slow system has been “secret data factories”. This directly inhibits the biggest criterion for a good BoM system: the elusive “one master BoM”.
In a more human-centric aspect, overly complex and slow systems have a cultural effect. One system we observed where loading BoM data took more than 10 minutes (due to high complexity of validations and checks) led to a culture of the users requesting BoM data and then going on to a coffee break. In a study, it was specifically labelled by the participating engineers as the “worst system to work with” and was directly related to driving down the efficiency of the business.
Striking the balance
Your data quality will directly be impacted by how well your system is balanced. Having understood the trade-off between speed and intelligence and considering your business needs, there are four main factors to consider when looking at your balance point:
Rate of data change: infrequent changes allow more leeway regarding the need for speed, whereas high rates of change can call for a faster system.
Size of BoM: large volumes of data can be quickly fed through the business in order to achieve desired efficiency. On the other hand, intelligently embedding your business process could help better manage great data volumes, but could make the task of interacting with the BoM a slow, painful process.
BoM complexity: high complexity increases the information that a system needs to process even with small BoMs. The trade-off between speed and intelligence is critical here as the user input needs to be efficient. So, do you choose quick access or complex checks to ensure more value-add information is provided?
Business process complexity: a relatively simple process will generate less data, even in a complex system, allowing you to move faster and more efficiently. On the other hand, high business process complexity could mean having data stuck in the labyrinth of corporate bureaucracy. An intelligent system could help you safeguard against such modes of failure.
At Quick Release, we believe that the system is not the whole answer. In order to deliver successful programmes, progressive data maturity is critical.
To enable this, a system must be fast enough to encourage users to keep their BoM up-to-date and avoid “secret data factories”, but also clever enough to enforce that data maturity. With the increasing complexity of data in a more connected and electrified world achieving this becomes even harder. A balanced system will contribute to success. However, to deliver this success in a truly effective manner, PDM professionals who are process and system experts are essential.
Originally published on Medium: https://medium.com/@quickrelease_/fast-vs-clever-bom-30fd35d13663