System dynamics

We have seen now that lead time reduction is a very important activity within QRM. To realise a huge reduction it is necessary to adapt your organisational structure. Besides it is very important to know something about system dynamics and to know how to handle this.systeemdynamica afbeelding 1 snelweg

The MCT is the result of interactions and dynamics between the available sources and tasks to perform. In a system that is full, you will get that the parts in the system start to influence each other on a given moment. They will call for a reaction. To understand how this works or even better how to avoid this is of the greatest importance to reduce the MCT. A highway with traffic on it is the ideal metaphor to explain system dynamics.

How do traffic jams originate in traffic? They originate through an increasing number of cars and an increasing variability. The number of cars seems easy, the more the road is utilised the sooner you get a traffic jam. But what is a variability? Variability are variations in for example driving speed, someone who breaks, someone who changes lanes, but also influences like rain, sunlight, trucks, junctions and so on. The more variables we have the sooner the chance for a traffic jam.

doorlooptijd variabilitteit grafiek engTo explain the relationship between lead time, utilisation and the variability, you can look at the figure. On the horizontal axis we have put the utilisation, the vertical axis gives us the lead time. When the traffic is quiet and the utilisation is low, the lead time will be relatively low (lower left site). When the traffic increases and the utilisation increases, at first the lead time will hardly change. With a further increase in traffic the lead time will start to increase also. However there is still little interaction. At a certain moment we will see the effects of system dynamics, sources start to interfere and the lead time will increase progressively. This is usually the area where the utilisation is higher than 70%. When you have a low variability, the effects on the lead time will be lower. The higher the variability the sooner the progressive increase in the lead time will take place. What we want to avoid is to have a process where we end up in the coloured area. That is the area where traffic jams occur and where you’re flow comes to a halt.

It is very difficult to name a number or percentage up to where we can utilise the process. The main importance is that we plan to a level where work keeps flowing in the process. A good measure therefore is flow, as long as the work keeps flowing we can add more work, and as soon as work starts to pile up we have to stop adding more. How can we make sure that the work remains flowing? There are three factors that can realise this:

Reduction of variability

Variability is the difference in arrival time and the difference in production time of the orders. How can we reduce the variability?

  • No selling campaigns any more that can cause peaks
  • No rebates for quantum sales
  • No more combining of orders during the week
  • Daily despatch of orders
  • Try to equalise the orders as much as possible with regard to lead time (= operation time + set-up time)

Increase your spare capacity

  • Take care of setup time reduction
  • Try to reduce the operation time
  • Take care of clear utilisation of the source and avoid losses like rework, rush jobs, producing-to-stock, machine stops etc.

Reduce the total time per order

efficiency batch grafiek engTo reduce the total time per order you need a totally different approach when you look at your lot size.

With QRM the lot size is not based on costs but on lead time. With the traditional lot size, based on efficiency, the lot size will be as big as possible. This choice however has a very negative effect on the flow, the bigger the lot size the longer the lead time. The lot size with QRM is determined in a different way, being as short as possible lead time. Take a look at the next figure:

omsteltijd seriegrootte grafiek engThe minimum of the lot size is very dependent on the set-up time. The more we can reduce the set uptime, the closer we can get to the lot size of one (One-piece-flow). Within QRM SMED(program from Lean to reduce the setup times) is a very important tool to reduce the lot sizes and with that reduce the lead times.

You will find more information about system dynamics in appendix D from the book ‘It’s about time’.