THEORY OF CONSTRAINTS AND SYNCHRONOUS

MANUFACTURING

Sessions 13, 14 & 15

Introduction

• The Theory of Constraints (TOC) was developed by Dr. Eli Goldratt to aid manufacturers schedule their production better and make better use of their resources and inventories.

• The “Five Focusing Steps of TOC” are condensed below:

• Identify the system constraint. (no improvement is possible unless the constraint or the weakest link is found)

• Decide how to exploit the system constraint. (Make the system as effective as possible).

• Subordinate everything else to that decision. (Align every other part of the system to support the constraint even if this reduces the efficiency of the non-constraint resources).

• Elevate the system constraints. (If output is still inadequate, acquire more of this resource so it is no longer a constraint).

• If in the previous steps, the constraints have been broken, go back to step1; do not let inertia become the system constraint.

Production Scheduling Rules

• His rules for production scheduling are listed as under:

• Do not balance the capacity – balance the flow.• The level of utilization of a non-bottleneck

resource is determined not by its own potential but by some other constraint in the system.

• Utilization and activation of a resource are not the same.

• An hour lost at a bottleneck is an hour lost.• An hour saved at a non-bottleneck is a mirage.

• Bottlenecks govern both throughput and inventory in the system.

• The transfer batch may not and many times should not be equal to the process batch.

• A process batch should be variable both along its route and in time.

• Priorities can be set only by examining the system’s constraints. Lead time is a derivative of the schedule.

• The above principles underlie the concept of synchronous manufacturing, which refers to the entire production process working in harmony to achieve the profit goal of the firm.

• According to Goldratt:

• “The goal of a firm is to make money”

Performance Measurements

• To adequately measure a firm’s performance, two sets of measurement s should be used –

• Financial & Operational.

• The financial measures are:• Net Profit – an absolute measurement in dollars• Return on Investment – a relative measure

based on investment• Cash flow – a survival measurement.

Operational metrics

• While financial measurements work well at the higher level, they cannot be used at the operational level.

• So here we have:• Throughput – the rate at which money is

generated by the system through sales.• Inventory – all the money that the system has

invested in purchasing things it intends to sell.• Operating expenses – all the money that the

system spends to turn inventory into throughput.

Throughput• Throughput as specifically defined as

goods sold. • An inventory of finished goods is not

throughput but inventory. Actual sales must occur.

• This definition prevents the system from continuing to produce under the illusion that goods might be sold. Such action simply increases cost, builds inventory and consumes cash.

Inventory

• Inventory that is carried is (in any form – WIP or finished goods) is valued only at the cost of the materials it contains. Labor costs and machine hours are ignored.

• Using just raw materials cost also avoids the problem of determining which costs are direct and which are indirect and their allocation.

Operating Expenses

• Operating expenses include production costs (such as direct labor, indirect labor, inventory carrying costs, equipment depreciation and materials and supplies used in production) and administrative costs.

• The key difference is that there is no need to separate direct and indirect costs.

• From the operational standpoint, the goal of a firm is to

• “INCREASE THROUGHPUT WHILE SIMULTANEOUSLY REDUCING INVENTORY AND OPERATING EXPENSES.

Productivity

• Typically, productivity is measured in terms of output per labor hour. However, this does not ensure that the firm will make money. The questions to be asked are:

• “Has the action increased throughput?”

• “Has it decreased inventory?”

• “Has it decreased operational expense?”

• This leads us to a new definition:

• “Productivity is all the actions that bring a company closer to its goals”

Capacity

• Typically, manufacturing tries to balance capacities across a sequence of processes in an attempt to match capacity with market demand.

• In synchronous manufacturing thinking, however, making all capacities the same is viewed as a bad decision.

• Such a balance would be possible only if the output times of all stations were constant or had a very narrow distribution.

• A normal variation in output times causes downstream stations to have idle time when the upstream stations take longer to process. Conversely, when upstream stations process in a shorter time, inventory builds up between stations. The effect of the statistical variation is cumulative.

• The only way this variation can be smoothed is by increasing wip to absorb the variation (a bad choice since we are trying to reduce wip) or increasing capacities downstream to be able to make up for the longer upstream times.

• The rule here is that capacities within the process sequence should not be balanced to the same levels. Rather attempts should be made to balance the flow of product through the system.

• When flow is balanced capacities are likely to be unbalanced.

Dependent Events & Statistical Fluctuations

• If a process flows from A to B to C to D, and each process must be completed before passing on to the next step, then B, C, and D are dependent events.

• The ability to start the next process is dependent on the preceding one.

• Statistical fluctuation refers to the normal variation about a mean or average.

• When statistical fluctuations occur in a dependent sequence without any inventory between workstations, there is no opportunity to achieve average output.

• When one process takes longer than the average, the next process cannot make up the time.

• Recall the two stage process simulation that we did with a batch of six and an individual process time of 10 minutes average and various standard deviations.

• In an eight hour shift the average output of the process was less than the 48 predicted by averages.

• It is such statistical variations that TOC tries to address by aiming at balanced flow through the system rather than balanced capacities.

Bottlenecks & Capacity Constrained Resources

• A bottleneck is defined as any resource whose capacity is less than the demand placed upon it. A bottleneck is a constraint within the system that limits output. It is that point in the manufacturing process where flow thins to a narrow stream. A bottleneck can be a machine, scarce or highly skilled labor or a specialized tool.

• A non-bottleneck, on the other hand, is any resource whose capacity is greater than the demand placed on it.

• A non-bottleneck, therefore, should not be working constantly because it can produce more than is needed. A non-bottleneck contains idle time.

Capacity-constrained Resource• A capacity-constrained resource (ccr) is one

whose utilization is close to capacity and could be a bottleneck if not scheduled properly.

• For example, a ccr may be receiving work in a job-shop environment from several sources.

• If these sources schedule their flow in a way that causes idle time for the ccr in excess of its unused capacity time, then the ccr becomes a bottleneck when the surge of work arrives at a later time.

• This can happen if batch sizes are changed or if one of the upstream operations is not working for some reason and does not feed enough work to the ccr.

Time Components

• The following kinds of time make up production cycle time:

• Set up time – the time that a part spends waiting for a resource to be set up to work on this same part.

• Processing time – the time that the part is being processed.

• Queue time – the time that a part waits for a resource while the resource is busy with something else

• Wait time – the time that the part waits not for a resource but for another part so that they can be assembled together.

• Idle time – the unused time; that is the cycle time less the sum of setup time, processing time, queue time, and wait time.

• Often, schedulers are tempted to save setup times – usually by increasing batch size.

• Suppose the batch size is doubled: • 1. Setup time will be saved by half – but• 2. All of the other times – processing, queue and

wait times – will double.• 3. The net result is that the work-in-process is

approximately doubled, as is the investment in inventory.

• The crucial step is to identify the bottleneck and add resources there to increase its capacity.

• An hour saved at the bottleneck adds an extra hour to the entire production system.

• An hour saved at a non-bottleneck is a mirage and only adds an hour to its idle time.

Drum, Buffer, Rope• In any system, the bottleneck is the best place to

control the flow of product through the system. This control point is called the drum because it strikes the beat that the rest of the system uses to function.

• By definition, a bottleneck is working all the time and one reason to use it as the control point is to make sure that the upstream operations do not overproduce and build up excess wip that the bottleneck cannot handle.

• If there is no bottleneck, the next best place to set the drum is would be a ccr.

The Bottleneck• Dealing with the bottleneck is the most critical

and the focus will be to ensure that the bottleneck always has work to do.

• Consider the simple linear flow from A to G. • Let D be the bottleneck. • This means that the capacities are greater

upstream and downstream to it. • There would be little finished goods inventory

because by the definition of the term bottleneck, all products produced would be taken by the market.

A B C D E F G

Market

Inventory Buffer

Communication (rope)

Bottleneck (drum)

• There are two things we must do with this bottleneck:

• Keep a buffer inventory in front of it to ensure that it always has something to work on. Since it is a bottleneck, its output determines the throughput of the system.

• Communicate back upstream to A what D has produced so that A provides only that amount. This keeps inventory from building up. This communication is called the rope.

Buffer Size

• How large should the buffer be?• Theoretically, the size of the buffer can be

computed statistically by examining past performance data or the sequence can be simulated.

• However, precision is not critical. • We could start with an estimate and observe the

buffer. If it runs out after some time, it means we need more buffer while if it builds up or remains high, we could reduce it to a more manageable level.

• If the market cannot take all that the process produces, we create a finished goods inventory buffer at the end of the line and a time buffer in front of the bottleneck/ccr.

• The time buffer keeps the bottleneck completely occupied and the finished goods inventory protects the market from a stock-out.

• In such a case we will require a rope from the finished goods inventory to the bottleneck to prevent over build-up of that inventory.

How TOC Morphs to Lean

• It will be noticed that the TOC when extended to all stages in a process becomes the kanban system.

• JIT is the method devised to keep the buffer inventory at each workstation at the minimum.

• Please note that by leveling production and reducing batch size, each stage starts behaving like a bottleneck or ccr!

Batch Sizes

• In an assembly line what is the batch size?

• It is “one” if we concentrate on the number of parts transferred from one station to the other or a part focus.

• If the focus is the process then it is infinity since it is continuing to run the same units.

• In other words, we have a process batch of infinity and a transfer batch of one.

• In the context of TOC, • The setup costs relate to the process batch

while • The carrying costs relate to the transfer

batch.• Larger batch sizes require fewer setups and can

therefore generate more processing time. And more output.

• For non-bottleneck processes, smaller process batches are desirable since they use up existing idle time thereby reducing wip inventory.

• The transition from a large batch at the bottleneck to smaller batches for the up and down stream non-bottleneck stages is achieved by the transfer batch which is usually of a smaller size than the process batch.

How to Treat Inventory

• Goldratt and Fox propose to treat inventory as a loan given to the manufacturing unit.

• The value of the loan is based only on the materials cost, without any accounting-type value added from production.

• The loan is measured in terms of dollar days.

What are dollar days? (or Rupee days)

• It is the product of the total value of inventory and the number of days it spends within the department.

• For example an average inventory of $40000 held for 5 days would give a dollar day value of 200,000 dollar days.

• This then becomes one of the performance measures of the department.

• It would automatically propel the department to reduce dollar days and thereby reduce wip, promoting faster flow through the system.

Comparison of Synchronous Manufacturing and JIT

• JIT is limited to repetitive manufacturing.• JIT requires a stable production level (usually

about a month long).• JIT does not allow very much flexibility in the

products produced. (products must be similar with a limited number of options)

• JIT still requires wip when used with kanban so there is something to “pull’. This means that completed work must be stored on the downstream side of each workstation to be pulled by the next station.

• Vendors need to be located nearby because the system depends on smaller, more frequent deliveries.

• For continual improvements to the system, JIT is a trial and error procedure applied to a real system.

• In synchronous manufacturing, the system can be programmed and simulated on a computer because the schedules are realistic and computer run time is small.

Illustrative Example

• Given below are the process flows for Products A, B, and C.

• These products sell for Rs.20 (A), Rs.25 (B) & Rs.30 (C) respectively.

• Two resources Resource X and Resource Y are used to produce A, B, & C with the process time given in minutes as shown on the diagram.

• Raw materials are needed at the process steps shown, with the cost in Rupees per unit of raw material. (One unit is used for each product)

• The market will take all that you can produce.

Resource X

Resource Y

Selling PriceRs. 20

Rs. 25 Rs. 30

A B C

X X X

Y Y Y

RMRe 1

RMRe 2

1 min/part

2 min/part

RMRe 2

RMRe 5

4 min/part

3 min/part

RMRe 5

RMRe 9

3 min/part

5 min/part

• Which product would you produce to maximize gross margin/unit?

• If sales personnel are paid on commission, which product or products would they sell and how many could they sell?

• Which and how many of the products should you produce to maximize gross profit for one week?

• From 3, how much gross profit would there be for the week?

Solution

• Maximizing gross margins per unit:• Product B will be produced

Product Selling Price (Rs)

RM Cost (Rs)

Gross margin (Rs)

A 20 3 17 B 25 7 18 C 30 14 16

Maximizing sales commission

• Sales personnel would sell the highest priced product, C (assuming they do not know the capacity limitations)

• Capacity for C, with Resource Y the constraint, is 12 parts/hour

• So, for a week at 8 hours a day and 7 days a week, total production would be = 12*8*7 = 672 units.

To maximize gross profit /week, we need to compare profits/hour for each product

Product Constraint Resource

Production time on resource (min)

Output/hour Gross margin Gross profit/hour

A Y 2 30 17 510 B X 4 15 18 270 C Y 5 12 16 192

• If the constraint resource were the same for all the products, our problem would be solved and our answer would be to produce as much A as possible.

• However, X is the constraint for B; so the optimum could be a combination of A & B.

• To test this we check the value of Y for every hour of producing B.

• This value is = (60/3)*18 = Rs.360

• Since this is less than Rs.510 for A, we produce only A.

• Gross profit for the week is 30*8*7*17 = Rs.85,680

• What would the decision be if B’s gross margin were to be Rs.30/unit?