Sunday, February 24, 2013

Focus and Leverage Part 189

In my last posting I demonstrated how efficiency could have a negative impact on your operation, but before moving on to a solution or a replacement for this metric, I want to look at the negative impact of using efficiency as a metric a bit further.  What I demonstrated in my last posting was what the process would look like after 100 minutes.  Let’s look at the impact of running this process for a full shift and then 3 full shifts.  And by the way, for you purists out there, these examples assume 100% uptime (i.e. no breakdowns, no breaks or lunches).

In the figure below we see the impact on the process by running at top speed (100 % efficiency) for full 8 hours.  Since Step 1 only takes 10 minutes to produce a part, in a full 8 hour shift it produces 48 parts and passes them on to Step 2.  Because Step 2’s capacity is only one part every 20 minutes, it can only process half of Step 1’s parts or 24 of them.  Steps 3 and 4 have the capacity to process all 24 of the parts produced by Step 2.  So after 1 shift we see that the accumulated WIP in front of Step 2 is 24 units while the finished goods inventory is also 24.
The efficiencies at both Steps 1 and 2 are 100% and according to cost accounting beliefs, are both doing their jobs.  But because the Operators at Steps 3 and 4 are well below 100% they are not doing their jobs and should probably be disciplined.  So as before, the overall system efficiency is hovering around 67%.  So, based upon what we’ve seen here, does it make any sense at all to continue running Step 1 at capacity?  Need more proof?  Let’s look at the results after 3 shifts of operating this way.
After running this process for three full shifts (i.e. 1,440 minutes), look what’s happened to the WIP levels!!  Step 1 produced 144 units during the 3 shifts, but Step 2 could only process half of them (72), so the WIP has exploded to 72 units.  We were able to ship 72 units, but believe it or not, WIP costs a company cash.  So again I ask you….does it make any sense at all to continue running Step 1 at its full capacity?

OK, so if efficiency isn’t such a great metric, then how should we appraise how well our overall process is performing?  Let’s review just the first three of Goldratt’s 5 Focusing Steps and see if we might be able to move closer to the answer:
1.  Identify the system constraint or which step is limiting achievement of the goal of the   process.
2.  Decide how to exploit the system constraint or how do we get more out of our constraint
3.  Subordinate everything else to the above decision or for our 4-step process, don’t outrun the constraint.

Let’s return again to our simple 4-step process.  In our process, Step 2 has been identified as the system constraint simply because it has the least amount of capacity to produce units.  If we wanted to improve the output of our process, we must improve the capacity of Step 2 by reducing its cycle time.  And if we wanted to reduce the overall WIP in our process, then we must run all four steps at the same rate as the constraint, even though the other three steps can all produce units at a higher rate.  In other words, we must subordinate them to the constraint.
Efficiency in the cost accounting world is considered a measure of productivity, so we need an equivalent measure to use.  In my blog, many times I have discussed the benefits of using Throughput Accounting (TA).  The basis for TA comes from three individual measurements:

1.  Throughput is defined as the rate at which a company generates money through sales.  This doesn’t mean that we produce product to sit in a warehouse, it means for sales.

2. Operating Expenses includes all expenses except totally variable costs, which in most cases are direct material costs.  Another way of saying this is all expenses incurred when turning inventory into throughput.  This includes all labor costs (i.e. both salaried and hourly labor costs).

3. Inventory includes all purchases that a company makes for things it intends to sell.

The key is that throughput isn’t considered throughput until money is received from the sale of products.  In looking at it this way, we are forced to consider the entire system rather than localized parts of the system.  So how do we measure productivity in the TOC world?  The way we do it in the TOC world is to divide throughput by operating expenses or T/OE.  For example, if each product sold for $100 and our OE was $25, then our productivity measure would be 4.0.  If we improved our throughput to $200 and our OE remained the same, our new level of productivity would be 8.0.

The natural managerial tendency for many companies that utilize traditional Cost Accounting (CA) is that when excess labor capacity is exposed there should be a labor reduction.  So when operators in non-constraints have no product to work on, like in our example process, in the CA world this is considered a bad thing and typically lay-offs result. However, in my way of thinking, if a company is truly interested in moving forward into a world-class competitive arena, lay-offs should be avoided at all costs.  I do realize that sometimes due to economic conditions that a company can’t avoid layoffs, but these times must not happen as a routine way of doing business.  Layoffs can clearly send the wrong message to the remaining employees about the effects of improvement programs and the improvement effort will come to a virtual standstill.

So what happens in companies that embrace TOC?  What happens when excess capacity is realized?  These companies do one of two things.  One thing they do is focus on developing new products and entering into new markets.  The problem with traditional cost accounting is that companies that make decisions based on cost accounting metrics view labor as a variable cost rather than a fixed cost.  If, for example, you keep the same labor in place, but you create new markets, the new revenue, minus totally variable costs (e.g. raw material costs, sales commissions, shipping costs), all flows directly to the bottom line!  Wouldn’t this be better than laying people off?  To me it clearly is!!

The other possibility is to increase market share of existing products by differentiating your company through things like superior on-time delivery, superior quality, and sometimes even price breaks into other market segments as long as these reductions don’t create a price war.  So how would you be able to increase your capacity to support additional shipments?  By focusing your improvement efforts on the constraint to drive throughput levels higher.

In my next posting, we’ll take a look at some other performance metrics and see how they impact our improvement efforts.

Bob Sproull


No comments: