The Theory of Constraints and Drum–Buffer–Rope
Drum – Buffer – Rope (DBR) is an operations scheduling methodology based on Dr Eli Goldratt’s Theory of Constraints (TOC) and first written about in The Goal and further explained in The Race. Drum Buffer Rope is just one part of the TOC Operations solution; it is the machine that sets the plan for Operations. However the second part of the TOC Operations solution is Buffer Management. Buffer Management is the monitor and control mechanism that ensures the machine is running well in execution.
Theory of Constraints takes a systemic view of a company or organisation. This means it treats the organisation as one and not as a number of different departments. An organisation is made up of a set of interdependent elements working together to achieve a company goal or objective. It is, therefore, important to relate all actions within an organisation to its impact on the global objectives. It is also important to realise that not all elements affect the system in the same way; some elements have less capacity than others. The elements with less capacity are the constraints of the system and limit the system’s ability to achieve its global objectives. If a (for profit) organisation had no constraints, then its profit would be infinite. In reality, there are very few constraints to any system.
The 5 Focusing Steps
Theory of Constraints uses this view to derive a 5 step process, called a ‘Process Of On-Going Improvement’ (POOGI), to improve any system. These steps are called the 5 Focusing Steps:
(Step 0 – Define the system’s goal or objective)
Step 1. Identify the System’s Constraint(s).
Step 2. Decide how to Exploit the System’s Constraint(s).
Step 3. Subordinate everything else to the above decisions.
Step 4. Elevate the System’s Constraint(s).
Step 5. If in the Previous Steps a Constraint has been Broken, go back to Step 1. (Warning: Do Not Allow Inertia to Cause a System Constraint)
The difficulty in Operations
In production, many different scheduling approaches are used to attempt to manage the capacity of the system and protect the system from the impact of variation (Murphy’s Law), so as to deliver the product or services “on time”.
The difficulty facing operations environments is to find a solution in environments where demand is fluctuating, Murphy strikes frequently, variation is everywhere and conditions rarely remain constant.
Just In Time (JIT) provides one solution to this problem. The JIT philosophy holds a minimum inventory between each work-center, typically only one. When there is a blockage in the flow line and Murphy ‘strikes’, the line stops and plant output is effected. This output can never be recovered by the system as in most cases the flow line is balanced.
Just-in-time does an excellent job of minimising inventory in the system thus increasing the flow and minimising the lead time. However it assumes variation in the process and capacities are minimal and focuses on reducing this variation everywhere.
Drum-Buffer-Rope enhances JIT by protecting the weakest link in the system, and therefore the system as a whole, against process dependency and variation and thus maximises the systems’ overall effectiveness. The outcome is a robust and dependable process that will allow us to produce more, with less inventory, less rework and defects, and better on-time delivery – always.
Applying the 5 Focusing Steps
Following step 1 – Identify the System’s Constraint(s), it is important to determine where the constraint in the system lies. With operational environments, the constraint can either be internal or external. If there is more market demand than the system capacity then there is an internal capacity constraint. If there is more capacity available but orders are not available to process then the system has an external market constraint.
For production companies with a true internal constraint Drum–Buffer-Rope (DBR) is the preferred solution. For production companies with an external market constraint Simplified Drum–Buffer–Rope (S-DBR) is the preferred solution.
However, many environments often appear initially to have an internal constraint as they do not currently have the capacity to process all the orders. Yet after the application of DBR or SDBR these environments have more capacity than the orders available to process so in reality they are externally constrained. It is the policies and rules currently used to plan and manage the environments that constrain operations not there capacity.
When there is an internal constraint, there are very few resources (people, machines, equipment, materials) dictating the output of the system. The most limiting resource is referred to as the ‘Drum’ as it determines the pace or ‘beat’ of the entire system. Following Step 2 - Decide how to Exploit the System’s Constraint(s), the constraint resource cannot be allowed to waste one moment of its capacity. This means that it should never be stopped waiting for parts and should not use capacity producing anything other than the parts required to fulfill sales orders. To ensure this we finitely schedule the Drum creating a Drum Schedule. The Drum schedule should maximise the Throughput of the Constraint and provide a detailed plan for just this one area. The Drum Schedule must be derived from the Shipping Schedule.
Following step 3 - Subordinate everything else to the above decisions, there are a number of actions that have to be met by the non-constraints in the system in order to meet the Drum Schedule and ultimately the shipping schedule. As discussed earlier, variation and Murphy cause, from time to time, pieces of plant to break down. Understanding that we have to protect the constraint from lost capacity due to these breakdowns, a Buffer of time is used. The Constraint buffer is a pre-determined length of time; we must release the order into the system before the order is due on the Drum Schedule. As all other resources have more capacity than the constraint (by definition), the effect of introducing parts a buffer time before they are due at the constraint is that work builds up in front of the constraint and protects it from breakdowns on preceding operations.
To ensure that too much inventory is not introduced into the system, it is important to start a new order only as the constraint finishes one. To ensure this, a Rope is tied to the first (gating) operation of the system. This is calculated by the date the order appears on the Drum Schedule minus the Constraint Buffer time giving a schedule for material release into the system. We have now “choked” the release of work into the system.
To ensure the Shipping Schedule is met after meeting the Drum Schedule a rope is tied from the Drum to the Shipping Schedule. The rope length is a buffer of time to ensure the shipping schedule is always met; this is called the shipping buffer.
This “choking” of release results in excess capacity being revealed in the non-constraints in front of and behind the Drum. This can have negative ramifications and must be properly handled in implementation; otherwise non-constraint resources will slow down to protect themselves from perceived negative consequences of not being busy all the time. This will of course impact Throughput and delivery performance.
When there is an external constraint, there is more capacity available than orders to fill it. Therefore, the Drum is set as the market. Following Step 2 - Decide how to Exploit the System’s Constraint(s), with enough capacity and limited volume of orders, no order should be delivered late. To ensure this, the system must be geared to delivering all customer orders - the Drum Schedule. This means that the Drum Schedule assumes infinite capacity in the system and hence the shipping schedule and the Drum Schedule are the same.
Following step 3 - Subordinate everything else to the above decisions, there are a number of actions that have to be met elsewhere in the system in order to meet the Drum Schedule. As discussed earlier, variation and Murphy cause, from time to time, pieces of plant to break down. Understanding that we have to protect the constraint from lost capacity due to these breakdowns, a Buffer of time is used. The Shipping Buffer is a pre-determined length of time; we must release the order into the system before the order is due on the Drum Schedule/ Shipping Schedule. This ensures that all orders have ‘enough’ time to be processed fully by the customer due date.
To ensure that too much inventory is not introduced into the system, a Rope is once again tied to the first (gating) operation of the system. This is calculated by the date on the Drum Schedule minus the shipping buffer time for each part giving a schedule for releasing material into the system.
There are some environments that do possess an internal constraint yet have products which do not pass through it. These are called “free products” and environments which have both constrained and free products must combine S-DBR and DBR.
The Machine is now set, however we must make sure that the machine is working properly and for this we need Buffer Management.
Buffers are used to protect against disruptions and to identify blockages very visibly and early enough in execution for recovery actions to be taken to ensure that the delivery commitment is still met. The information is collected on the reasons for delays. This is analysed, and focused actions are put in place to address them, continually improving the execution process and driving the lead time down.
The Buffers, the Constraint Buffer and the Shipping Buffers are effectively divided into three equal time periods. For example if the Shipping Buffer was 12 days then each of the three zones would be four days long. If the order was nine days away from the date required on the shipping schedule it would be in the first zone which is coloured green. If the order was five days away from the date required on the shipping scheduled then the order would be in the second zone and coloured yellow. If the order was two days away from the date required on the shipping schedule then the order would be coloured red. Buffer Management then becomes the priority system on the shop floor. A resource must always work upon a red in preference to a yellow or green and upon a yellow in preference to a green.
In order to ensure the red orders are expedited through the system and delivered on time Buffer Meetings are held to focus everyone’s time and action on the very few important things in the system. These meetings have all relevant staff and are held for 10 minutes each day or shift. Actions from the previous days meeting are checked to ensure they were completed and actions assigned to the relevant people for today’s reds.
Objectives of Meetings
- Provide the right information to the relevant people to enable them to quickly make the correct decision operationally now for the business
- Provide the right information to the relevant people to enable them to focus on the improvements to the business in the future that will have the biggest impact on Throughput, Investment and Operating Expense
- Communicate decisions and the reasons for making them to the other parties to enable proper dissemination to their peers and sub-ordinates
The major decisions to be made in a manufacturing environment are on:
- Managing the flow of work from Raw Material to Dispatch
- Managing the quality of the parts to the customer
- Managing the Sales, Throughput (T), Investment (I) and Operating Expense (OE) internally
- Managing the flow of work from Raw Material to Dispatch
Generally, the material flow can be affected by the following:
- Raw Material availability
- WIP material (size of the queue)
- Availability (Health & Safety, Sickness, staff turnover)
- Preventative Maintenance & Breakdowns
- Subordination (bottleneck busting)
Buffer Management provides the right information to the relevant people to enable them to quickly make the correct decision operationally now for the business on all of the above.
The following meetings are held daily or per shift to ensure that the decisions and actions are taken to increase the flow of the orders through the system.
There is an assumption the Operations department can only impact the sales of the company by delivering them all on time in a short lead time. It is Sales and Marketing functions responsibility to provide Operations with the sales and to leverage Operations excellent performance. For a short period of time the Op