TPS or Toyota Production System is world renowned in manufacturing as being at the pinnacle of best practise in production control.  Many of the lean principles that are used today were developed and pioneered by Toyota.

TPS has three desired outcomes; firstly, to provide the customer with the highest quality vehicles, at lowest possible cost, in a timely manner with the shortest possible lead times. Secondly to provide members with work satisfaction, job security and fair treatment. Finally, it gives the company flexibility to respond to the market, achieve profit through cost reduction activities and long-term prosperity.

TPS strives for the absolute elimination of waste, overburden and unevenness in all areas to allow members to work smoothly and efficiently. The foundations of TPS are built on standardisation to ensure a safe method of operation and a consistent approach to quality. Toyota members seek to continually improve their standard processes and procedures in order to ensure maximum quality, improve efficiency and eliminate waste. This is known as kaizen and is applied to every sphere of the company's activities.

In an automotive assembly plant like Burnaston, component parts are made available to the assembly line from small kanban controlled WIP stores. To balance the assembly line the product mix, sequence and repetitive period is typically pre-defined each month (such as 10 Type A, 10 Type B, 10 Type A, 10 Type B, repeating every day). Each WIP store will contain a number of 'kanbans' or 'totes' of the components and each time one tote is removed from the store a kanban signal is sent to the component production cell as an 'order' to replenish the store. The minimum production quantity is one tote full.

In a perfect world the component cells would simply make one tote full of each item whenever a kanban signal arrives, but the world is far from perfect and the component cell may have to develop a schedule that is very different to that of the main assembly line.

For example, the dashboard injection moulding cell at Toyota's UK plant cannot make all the components one tote at a time because the number of setups required means that they would not have enough capacity to meet their demand. They have to make larger batches, but at the same time they must keep the WIP stores between their minimum and maximum levels. The problem is further complicated by the fact that the component cell works three shifts to achieve the required capacity whereas the main assembly line works two shifts.

Calculating a feasible sequence of component production to match the demand, given the parameters of the demand pattern, tote size, minimum and maximum store levels, production rates, setup times and shift patterns is known as Heijunka (load-levelling). This is a non-trivial calculation that must be carried out every time the assembly mix, quantity or sequence varies.

A Preactor Advanced Planning and Scheduling system with a “Heijunka” scheduling rule is being piloted by Toyota to calculate the component sequences and batch sizes, and has dramatically reduced the time required to perform the calculations.

Many lean manufacturers would see the level monthly demand typically achieved by the automotive sector as unobtainable, and the more variable the demand, the more Heijunka scheduling is required to assist the kanban production control.

Finally, if we extend the concept of Heijunka scheduling across more component cells and the final assembly areas of our lean factory we can achieve visibility of the effects of the variable demand leading to better decision making about priorities, delivery performance, resourcing levels, etc.