Dear Kidanemariam -

By "firm," I assume you mean a business entity, and I am not familiar with your data, but perhaps I have seen something analogous: For electric power generators, I think that a machine is stamped with a "nameplate" capacity, which is the (nominal) maximum generation one might expect it to produce, considering the specifications under which that equipment was manufactured.  Sometimes, I think, such a generator may be called upon and able to exceed this. 

It would appear that in your data there are cases where some level of performance is generally considered to be peak performance, but that that is not necessarily the case. 

In any such situation I would think that true maximum performance is not easily pre-determined. 

Cheers - Jim

Hi James, I guess you formulated an output oriented model.

Let assume you had a score of 1.5...It can be interpreted as "the DMU under consideration must increase it output levels to 1.5 times its observed (actual/current) levels in order to become efficient". This means that the particular firm has capacity for output maximization and is therefore inefficient.

Alternatively,  you can find the inverse of the output efficiency score (Farrel efficiency). This will give you a Shephard distance function value of 0.67 (i.e. 1/1.5). This can be interpreted as : "The firm is currently production at 67% of its potential output levels (or its potential capacity)". Since the firm is not producing at 100% of its potential capacity it is inefficient.

I hope this helps

Charles

Another point to note, there is a different interpretation for input oriented models. If this is what you are relying on, then let me know.

There is also what is referred to as "super-efficiency". This is normally seen when a DMU is a separate frontier is measured relative to the production technology that it does not belong to. So, if you do..Malmquist productivity analysis or Global frontier analysis, you'll see many such situations there.

Charles talks about models. Kidanemariam wants to know how the over the 100% capacity production is possible. 

Normally a firm or more precisely a factory or a production site determine its production capacity as the output that is possible by using the actual machines and installations, actual labor power within the normal labor time, for example 8 hours a day. The full capacity is not the maximum of capacity that is possible under the present situation. If demand  comes more than usual, the factory or production site can produce than the full capacity by (1) prolonging the working time or by (2) increasing the labor power  temporarily.  If the normal production is made in one shift system, it is quite easy to extend its production rate 100% for example.

There are, of course, other kind of production process. In the case of iron and steel industry, pig iron is produced 24 hours a day. It is because one cannot stop the high furnace. High furnace runs many weeks if once fire was lit. In this case, full capacity is defined as normal produce by this continuous production. However, even in this case, over capacity production is possible. I once visited a steel maker at a time when the demand was quite large. At that time they set supplementary gas burners at the bottom of the furnace and in this way increased the heat necessary to melt and reduce the iron ore more rapidly. In this case. the production capacity increases but the unit cost may increase.  Therefore overcapacity production is not necessarily the most efficient production if we use the term "efficiency" properly.

The above is knowledge that everybody economists should know, but in reality a few people know this. Economics textbooks talk much about production function, but  do not explain how the real production is organized. This is only a tip of huge iceberg which is hidden by neoclassical textbooks. If once students come to know the real production process, they cannot accept such notions as smooth production functions. Kidanemariam asked a very good question. If one tries to consider with common sense, we come to know the fictive characteristics of the neoclassical economics. 

Explanation to this depends much on the model employed!