In the world of academia, there are several theories to describe throughput. But for the common businessman, the only law that he needs to truly understand is turning production into maximum profit.
The following sections describe the various theories, laws and methods about throughput. The next area will illustrate how to identify throughput issues in an easy to understand format for the small business entrepreneur.
Theory of Constraints
For the layman, the theory of constraints is simply stated as the system is no stronger than the weakest link or process within the system. The key is to identify the weakest link and make it stronger. Keep identifying the weakest link and make continuous improvements to the system. This is the management philosophy for Theory of Constraints. For a really detail explanation, read: Dr. John Blackstone’s Theory of Constraints.
Within the theory is a measurement variation known as throughput. It is nothing more than the ability to generate maximum revenue. The question is: What is the weakest link in generating those sales? Is it production, the sales department, delivery or could it possibly be inventory. For the small business entrepreneur, find the weakest link and fix it!
In every business no matter production or service, there exists a bottleneck. Your job as the owner or manager is to find it and fix the problem. Once this one is found and resolved; move onto the next one. There is always a bottleneck to find and fix. What is important is to find the problem, identify the core issue, and then resolve the bottleneck. This may sound simple, but it takes someone in a position of knowledge and experience to complete this task. As the owner and or manager of the business, you are the in best person to act and address the bottleneck. Notice how much similar this is to the Theory of Constraints. I have written another article on this subject: Find the Bottleneck.
A form of cost accounting that is used as a performance measuring system. It is designed to provide the profitability of various combinations given available resources. The key is to produce outcomes that maximize cash given the limitations. Resources include raw materials, labor and capital costs associated with product production. This will be articulated in detail in a future article, for now, suffice it to say that this term in used to maximize resources and earnings.
In modern bank operations, when you go to the bank and you stand in line, you stand in a group line that feeds individual tellers. Back in the 70’s and 80’s, you chose your teller line and stood in that particular line. Invariably, one of the other tellers was either faster or had simpler transactions to handle. The guy you walked in with got faster service, you felt cheated. This is because of the form of the system. In the olden days, the system was just the individual teller, now the system uses all the tellers as a group to equalize the wait time as much as possible for all customers. It makes the average customer think that his wait time is no more than the next guy. In reality, it wasn’t any different back in the 70’s or 80’s. Over a course of 100 bank visits, I can assure you that your average wait time was equal to the overall group. This is the premise of Little’s Law.
Little’s Law is defined in units per time period. Simply stated it equals throughput times flow time to identify the inventory. In the above example, the folks standing in line are the inventory to manage. If each teller has an equal ability to handle the transactions, then the flow time equals inventory divided by the throughput (a teller’s average time to handle a customer). As an example, if there are 9 people in line and 3 tellers and each teller can process a customer in 2 minutes, than the maximum wait time (flow time) for a customer equals 6 minutes ((9 customers/3 stations which equals 3 customers per station) times 2 minutes per customer equals 6 minutes of flow time. Going back to the formula:
Inventory (number of customers in line) = throughput of 3 tellers handling 3 customers each times flow time of 2 minutes each. Therefore, inventory = 3 tellers X 3 customers each X 2 minutes of work/customer or a total 18 minutes of backlog or inventory. So if you are the 9th customer in line, you are looking at a 6 minute wait to get served (18 minutes of backlog divided by 3 tellers equals 6 minutes of flow time). For the first customer in line, you will be served immediately as will customers two and three; the fourth customer will have to wait at least 2 minutes to start service as will customers five and six. Customers seven, eight and nine will have to wait at least 4 minutes to reach a teller and 6 minutes to get their service completed.
Instead of using customers, think of a production line making loaves of bread. If there are only so many ovens and each oven takes a certain number of minutes to bake a loaf, then how many loaves of unbaked bread should we have in the queue for the ovens to manage them. This is important to know as an unbaked loaf has a certain life expectancy before going bad etc. It is possible to allow the loaf to rise while it sits in the queue before baking. So in production, understanding Little’s Law eliminates waste and maximizes efficiency.
Now realize from above, there is a lot more to this than just this simple formula. You and I both know that the tellers are capable of handling customers at different rates within a finite period of time. The key is the overall average over hundreds of transactions to determine a teller’s respective flow rate. Each will be different in their flow rate and then the inventory presents an issue too. Some customers only want to deposit a check, others are businesses making their deposit and getting their respective change etc. for the next few days. Some customers are there to cash out bonds, which takes a lot of time to complete. So Little’s Law is easy if every transaction were the same and the tellers could process the transactions in the same amount of time. This is known as consistent units (transaction type, teller throughput rate). Once the variables begin to change or be inconsistent, the Law becomes a group dynamic measurement more than a single individual measurement in order to determine the outcomes associated with the respective units per measurement of time.
You can see how this is important in grocery stores as it relates to cashiers checking customers out, or in the fast food industry getting hot food to a customer as fast as possible. It is important to maximize the flow rate in order to maximize revenues and ultimately profit.
Throughput Issue Identification
To best demonstrate and illustrate the process, I’ll use a common business scenario. For this example I selected a restaurant operation. So imagine a family owned pizza shop that seats customers and has carry-out and delivery services. I added these two variables to illustrate the difference in the thought processes associated with throughput due to staffing costs.
Because variables really complicate or extend the identification process, I will keep the restaurant basic with the following variables:
- The only food product is a large pepperoni pizza
- The number of tables are 10 – 2 person tables and 18 – four person tables
- 2 full time wait staff, 3 part time for the weekends
- 1 delivery driver
In the restaurant business, the key is to fill the tables every night and try to get a high turnover rate on the tables, i.e. 3 seating’s per night. Of course, we are all smart enough to know this isn’t going to happen Monday through Thursday unless there is some other mechanism to attract more customers. However, on Friday and Saturday, the goal should be generating the highest turnover rate as possible to maximize revenues and ultimately the associated margins. Remember the goal is to get maximum production that will ultimately turn into profit.
So let’s start out with the Theory of Constraints to help us determine the best course of action. The theory of constraints tries to identify any variables reducing the ability to maximize throughput. So I’ll start with the food. Naturally, if all you offer is pepperoni, then some potential customers will not visit the restaurant. I’m an example of this. I like pepperoni but I will not select that flavored pizza unless there is no other choice. My wife and I are fans of sausage and spinach. She got me hooked on that flavor and so that’s what we prefer. So right out the gate here, we have a problem. As you add more flavors to the menu, you begin to expand the market of potential customers. At the same time you begin to complicate the operations of the restaurant. Keeping the supplies associated with the other ingredients adds storage issues, refrigeration issues and production complications. Think about how difficult it is to train a pizza maker when you add a variety of meats and vegetables to the menu. How much does he put on the pizza per item; how about those times when the customer requests half and half? At some point, you have to stop adding items to the menu. How many more customers are you going to attract to the store if you add zucchini to the menu? So there is a happy medium for this situation. As a restaurant owner, you sometimes have to adapt to the fads out there. An example is barbecue pizza. Really? How long is this going to last in the market of pizzas? But if it is the rage, then for a period of time, you will have to offer this flavor.
Another example of the theory of constraints usage deals with the turnover rate of tables. So how do you get the tables to turn over faster allowing you to serve more customers? Well, one key is to get the food on the table as soon as possible so those seated focus on eating and then they’ll feel the urge to depart. If no food on the table, they will not feel the urge to leave as they have yet to eat. Now this is easy if only one flavored pizza to choose from. Think about it for a moment, all the waitress asks is how many do you want? If the maker and ovens can keep up with the demand, no problem at all, so you should be able to maximize the throughput inside the store. There could be one other slow aspect of this and that is the server. Maybe one of the waitresses is slow or lacks attention to getting it done. This too will reduce the turnover rate. However, in the real world, the problem is waiting on the customer to order after reviewing the menu and then making the custom orders of pizza. So here the theory of constraints tells us that due to the delay in receiving the order and making the custom pizza, the turnover rate is going down. As you add more complexity to the process, the turnover rate continues to decrease. Examples include having to cut some vegetables if you run out of certain ones; not enough oven space to cook the pizzas based on the demand; busboy failing to clean the tables in a timely manner upon the patron’s departure; insufficient pre-made pizza dough and the list can go on.
So as a busy owner, you use the theory of constraints to identify relevant issues that will most likely cause or slow the turnover rate. This leads us to identifying the actual culprit. Here we use the ‘Finding the Bottleneck’ in the production line.
The production line begins with the order, whether by phone or by the table. An order is received in the kitchen and the pizza makers begin the process. Now if all we made were one size pepperoni pizzas, it will be easy for the maker. Just keep making pizzas until there begins a backlog of pizzas not headed out for consumption. If there is staff waiting on the pizza maker for orders, then it is obvious that either the pizza maker isn’t going fast enough or if he is; you need another pizza maker. Now let’s say we have two pizza makers going as fast as they can, and we still have staff waiting to fill orders. Then look at the production line, the oven isn’t getting the pizzas out as fast as the demand and there are pizzas waiting to go into the oven. This means you that we have an oven issue; there is a bottleneck at the oven.
The key to this term is finding the bottleneck. Believe it or not, there is always a bottleneck in production. You just need to keep hunting for the bottleneck. Sometimes though you run into a situation of getting the pizzas to the customers and it isn’t necessarily the production process. So let’s take a look at the issue as it relates to the delivery process. Notice that we have three ways to deliver, table service, carry-out, and delivery. So let’s walk through some math which is what throughput accounting is about.
To calculate the throughput accounting, let’s makes some assumptions:
1. Table service can deliver up to 40 pizzas an hour at a labor cost of $32 per hour for the waitresses.
2. Carry-out can deliver up to 60 pizzas an hour at a cost of $12/hour for the cashier.
3. Delivery can get 7 pizzas an hour out to patrons at a cost of $22/hour including the fuel.
Based on the above, you can see that carry-out is the most cost effective method of deliver at 20 cents per pizza. Table service costs 80 cents per pizza and delivery runs $3.14 per pizza. Ouch, that is expensive. So why on earth would you deliver pizza? Well, the answer is addressed via the production issue. If making pizzas is not a problem and other variables cause no delay in the turnover rate of tables, then the answer lies in the marginal contribution margin a delivered pizza generates. If each pizza generates a contribution margin (See Definition of Contribution Margin for more information about contribution margin) of more than $3.14 then it is to the pizza shop owner’s advantage to deliver pizzas. This is only true due to the ability to add marginal dollars to the pool. In addition, the pizza shop owner employs another individual and is providing employment.
In throughput accounting, the owner should focus on the most profitable form of deliver to the least profitable delivery of pizza. Here, carry-out is the most profitable than table delivery and finally home delivery. This is important to understand, if he hires a driver to deliver pizzas, then he really needs to make sure that home delivery is utilized to the maximum because any reduction in that quantity of delivery reduces the overall contribution margin quickly. Imagine only delivering four an hour at a cost of $22/hour. The margin in the pizza may not be enough to cover the additional costs of home delivery. So be safe, the pizza shop owner may want to consider a delivery fee similar to what Papa John’s charges for delivery. Sometimes, it may be best to limit the delivery periods to just Friday and Saturday night in order to ensure adequate requests for home delivery to cover the costs associated with that form of pizza delivery.
Now let’s move onto Little’s Law. Here the question is the inventory (not the food inventory). In the examples illustrated in the Business Description section above, the customer is the inventory. So in this example, we have the same situation. Here, there is a line outside trying to get seated to eat. Some of the inventory arrive in couples, others arrive as a family, sometimes more than four which means we have to use two tables to seat a family of five. So now based on this, you can see that we have two problems to deal with: first, there is a lack full utility of all the tables, i.e. if we could seat 92 patrons, we would want to seat all 92. But because of the variables in the way patrons show up at the door, this is highly unlikely and actually we’ll be lucky if we can utilize 80 seats of the 92 available. Little’s Law tells us that since we are using 2 full time waitresses and 3 part time waitresses that each waitress will have a different ability to process the patrons they respectfully handle. Of the five waitresses, one of them is going to set the standard to process patrons through the restaurant.
Little’s Law tells us to usher the next set of patrons to the most available table and corresponding waitress. This will minimize the wait time for those in line. Now some patrons will show up and will desire to be served by a certain waitress. Now their wait time will increase due to this restriction, it will also get complicated if a table size requested is unavailable. So you can see the value of processing information via Little’s Law to serve as many patrons as possible.
As you can tell the flow rate is different per waitress. However, if no server requests are made by the patrons, than the flow rate of the group is determined by the overall turnover rate of the entire wait staff. As an owner, if there continues to be a line of patrons outside and you have maximized the turnover rate within restaurant and there are really no bottlenecks or other constraints than he may desire to consider increasing the number of tables and corresponding wait staff. But this issue is for another article and is actually a good problem to have.
This example is a relatively simple example. It gets complicated when you add in issues associated with liquid refreshments and other menu items. Imagine the various bottlenecks associated with adding sandwiches, deserts and/or sizes of pizza.
The key to the above is the thinking process involved. The above example looked at the entire process of getting a pizza to the customer. Note the tools and methods used to analyze the overall situation. The Theory of Constraints teaches us to look for the weakest link in the system and fix that one first and keep looking for a weak link in getting the pizza to the customer. Bottlenecks hold up actual product getting delivered to the customer; so it is essential to identify the production issues. In the example above, it could be too much complexity in the customized pizzas, to insufficient oven space or a lack of precut ingredients. Similar to the theory of constraints, look for the production issue to address in getting the product delivered to the customer. Throughput accounting teaches us that at times we may need to look at the picture in a financial way and using tools like contribution margin and production volume can help us calculate the best to the least costs associated with maximizing production and ultimately profit.
Finally, Little’s Law is a tool to help us get as many customers through the system in the least amount of time per customer. This generates a better overall experience for the customer and maximizes throughput for the business.
By combining the above a small business owner can develop a good throughput model in his/her business operation. This in turn maximizes revenue per dollar cost and ultimately bottom line profit. Act on Knowledge.
If you have any comments or questions, e-mail me at dave (insert the usual ‘at’ symbol) businessecon.org. I would love to hear from you. If interested in my help as an accountant or consultant, contact me through the ‘My Services’ page in the footer.
If you found this article helpful, please consider a donation to the site. The donation button is just to the right. Even if you don’t make a contribution, I encourage you to read more articles on the website to help you become a better business entrepreneur.