All Laboratories who went through a lean lab exercise, see two prerequisites to become lean:
- Define standardized work for test methods, which requires correct lead times and workloads.
- Install large (magnet) boards for planning and visual management.
However, in the traditional lean lab approach, the promised gains come from basic tactics, such as:
- Create a flow to level demand
- Simple sample grouping rules
- Kanban
- …
There are many blends of lean. And – as every lab is a unique combination of workload, volatility, lead-time, equipment, and test types – the lean lab tactics will vary significantly from lab to lab.
The Achilles heel of the traditional lean lab approach is that it depends heavily on often counter-intuitive theory and research provided by (external) experts.
With an increasing variety of lab activities and demand, the tactics become too sophisticated. You’d have to be an excellent chess player to understand all the consequences and side-effects of your tactical decisions. In the end, Lean works best when there is 1 test method and 1 channel of demand.
But as lab circumstances change and expertise go elsewhere, … not well-understood tactics are not used anymore. Far too often, this results in lab schedulers spending a lot of time planning & manually juggling tests and campaigns, only to see sub-optimal results.
The solution is to fully harness the potential of modern computer algorithms (=digital lean lab). The benefit here: there is no need for everyone to understand how it’s done because they know they’ll be following the best path possible.
For us, the ‘digital lean lab’ should go further than just digitalizing the magnet whiteboard. It needs to make sure that:
- You no longer need to rely on experts and time-consuming projects to decide upon the best strategy. You need to be able to harness the full potential of modern computer algorithms to automatically calculate the best strategies time and again.
- Everything is constantly optimized: from grouping and holding samples to automatically optimizing the schedule based on due dates, priority rules, competencies, availabilities, and preferences.
How would this look like in your daily work environment? Well, we start from the same lean lab principles: Standard test methods (configured in the system) and a digitalized planning board (replacing the magnet whiteboard). Using readable decision trees (which is basic machine learning), Binocs balances sustainable productivity gains with improved service level (= planning adherence). A couple of screenshots:
Upload or configure your lean Lab timings & ways of working
