Enhance your sales process analysis for operations

How to create outlook signature

[Music] this video will provide an overview of the key concepts included in chapter 11 process analysis and resource utilization excellent customer service is a function of good day-to-day operations and process selection design improvement and analysis however providing excellent customer service comes at a cost since various resources must be employed in order to provide that service and if resources employed are underutilized then profits are eroded utilization is the fraction of time a workstation or individual is busy over the long run while high rates of utilization are desirable to reduce the amount of unused or idle time achieving high utilization is challenging and utilization will vary by process type and by industry there are two ways to calculate resource utilization the first simply takes resources used divided by resources available for example if we use 75 labor hours per week to finish shelving in a mail workshop and 100 hours are available then the utilization is 75 over 100 or 75 percent alternatively we can also calculate utilization to be the demand rate divided by the service rate times the number of servers for example if a call center answers an average of 100 calls per hour and we employ seven call representatives or servers that can answer 20 calls per hour each then the labor utilization is 100 divided by 140 that's 20 times 7 which equals 71 percent resource utilization is directly impacted by the efficiency of a process efficient processes allow for smooth throughput and have few bottlenecks throughput is simply the number of units or tasks that are completed per unit time from a process this could be measured as parts per day transactions per minute or customers per hour throughput is impacted by bottlenecks a bottleneck is the work activity that effectively limits the throughput of the entire process for example in a mcdonald's drive-through the payment window is a bottleneck as would be two service lanes merging into one bottlenecks often result in waiting lines or queues at any moment people orders jobs documents money or other entities that flow through the process are in various stages of completion and may be waiting in queues flow time or cycle time is the average time it takes to complete one cycle of a process for example it might take two minutes from order to presentation for a custer to make it through a mcdonald's drive-through it makes sense that the flow time through a drive-through will depend not only on the actual time to perform the tasks such as order payment and presentation but also on how many other customers there are in the system which we can call the work in process stage little's law is a simple equation that explains the relationship among flow time t throughput r and work in process or whip where whip is equal to r times t for example if a drive through serves an average of 75 customers per hour and it takes an average of two minutes to get through the drive through then whip equals 75 times 2 minutes divided by 60 minutes per hour which equals an average of 2.5 customers or cars waiting in line at any given time but who likes waiting in line nobody does but they're inevitable so companies put a lot of effort into managing waiting lines another term for a waiting line is a queue a queuing system consists of customers that arrive for service one or more servers that provide the service and a queue or waiting line of entities or people or customers that wait for service if the server is busy a drive-through is a perfect example of a queuing system so is a bank or supermarket customers servers and queues in a queuing system can be arranged in various ways and there are three common configurations the first configuration is where one or more parallel servers are fed by a single queue here's an example of what that looks like i guarantee you you've been in one of these whether you've been inside a bank or at inside a mcdonald's the second configuration is several parallel servers fed by their own queues does that look familiar it should be if you've ever been to a supermarket finally there's a combination of several cues in series that look like this a prime example is a voting station or even a passport office where there might be a pre-screening queue and then the final cue for processing managing waiting lines is serious business so there's an entire theory behind it queuing theory is the analytical study of waiting lines and this is where we can obtain some useful information to help manage lines better there are seven typical performance measures that can be calculated using queuing theory and are used to help improve the queuing system they include the probability that the system is empty i.e the probability of zero units in both the q and service the average number of units waiting for service in the queue the average number of units in the system the average time a unit spends waiting for service the average time a unit spends in the system the probability that an arriving unit has to wait for service and the probability of n units in the system analyzing queuing systems can be performed with analytical models or simulation models analytical models are simpler to use and can provide good estimates of the average long run behavior of queuing systems there are two basic queuing models each with its own set of key assumptions the first of which is called a single server queuing model and it assumes one that the waiting line has a single server two the pattern of arrivals follows a poisson probability distribution three the service times follow an exponential probability distribution four the queue discipline is first come first server or fcfs and five arriving customers must join the queue and cannot leave while waiting the second analytical model is the multiple server queuing model which has seven key assumptions the first is that the waiting line has two or more identical servers that serve customers from a single queue the arrivals follow a poisson probability distribution the service times have an exponential distribution the mean service rate is the same for each server the arrivals wait in a single line and then move to the first open server for service the queuing discipline is first come first serve and arriving customers must join the queue and cannot leave while waiting you can see some of the assumptions are similar as a single server model obviously in any queueing model servers cost money so we can calculate the total cost of waiting for service as a function of the number of servers the basic formula for calculating the total cost of weighting or tcw is csk plus cwl where cw is the waiting cost per hour per customer cs is the hourly cost associated with each server l is the number of customers in the system and k is the number of servers for most businesses the goal is to maximize throughput thereby maximizing cash flows now based on what we covered so far we know that throughput is dependent on various activities which may create bottlenecks in the system these bottlenecks represent constraints on the system and that now leads us to a short discussion on the theory of constraints or toc the theory of constraints is a set of principles that focuses on increasing total process throughput by maximizing the utilization of all bottleneck work activities and workstations a constraint is anything in an organization that limits it from moving towards or achieving its goal there are two basic types of constraints physical and non-physical a physical constraint is associated with the capacity of a resource such as machine employer workstation physical constraints result in process bottlenecks where the input exceeds the capacity restricting the total output that is capable of being produced imagine a three-lane highway suddenly being reduced to one lane only the bottleneck workstations are critical to achieving process and factory schedules and should be scheduled first time lost here represents time loss for the entire process and is very costly some ways to manage bottlenecks include using work in process as buffer inventory placed in front of the bottleneck to maximize resource utilization in addition larger order sizes can be employed at bottleneck workstations to minimize setup time and maximize resource allocation basically bottleneck should be producing at all times to maximize throughput and resource utilization a non-bottleneck work activity is one where idle capacity exists here idle time is acceptable if there's no work but resource utilization is typically low time lost at a bottleneck typically has no effect on total process or factory output and incurs no significant cost a way to manage non-bottleneck activities is to move jobs through them as fast as possible until the job reaches the bottleneck or small sizes can be used to keep work flowing to the bottleneck resources finally a non-physical constraint is environmental or organizational constraints such as low product demand or inefficient management policy procedures inflexible work schedules inadequate labor skills and poor management are all forms of constraints and unfortunately removing non-physical constraints may not always be possible great customer service does not happen by chance it is built into the very fabric of job process and supply chain design and improvement businesses put great effort and thought into effectively managing and utilizing resources to provide optimal customer service while minimizing costs and maximizing profits