This paper explores how companies have implemented Business forecasting to make informed decisions about the future of the company. Throughout the years this strategy has become extremely popular because it can help a company reach their goals by predicting short and long-term performance. There are various methods of such plan that can help a company in many ways. The methods can process small to a large amount of information. Below we will study three types of organizations that have implemented the use of forecasting methods to increase performance by making informed decisions.
One of the stores that Business forecasting has helped is Krispy Kreme, a retail store that provides sweets and coffee since 1937. Its operations began in Winston-Salem, NC… its founder Vernon Carver Rudolph born in Kentucky had one purpose, to build a doughnut empire. His vision started to come true in the early 1940’s-1950’s, at first it began as a small family owned and operated store, but then throughout the years it kept on expanding, consumers loved the freshness and consistency of the doughnuts.
When Krispy Kreme decided to go Public in the early 2000s, they also expanded into foreign markets. Today there are about 1,000 stores in 24 countries.
One of the aspects that diversifies Krispy Kreme from other doughnuts shops is the fact that they are consistent with their product, their doughnuts have a shelf life of 24 hours, this means each day they make fresh doughnuts from scratch. But, with a shelf life of 24 hours how do they know how many doughnuts to create for the day without wasting? How did they keep track to ensure that they are not creating unnecessary waste? Unfortunately, they couldn’t add a barcode to keep accounted for the doughnuts as retailers do with clothes.
At first, they began forecasting by using an in-house excel sheet, the problem with this is that calculations were incorrect, the process was time-consuming and unreliable.
Before Krispy Kreme decided to go public, they knew it was time for a change. They decided to do some research and look for a forecasting system that would provide information related to production and was based on demand-forecasting. A list of five steps was developed to find the ideal forecasting model strategy per Brad Wall the director of forecasting “the five-evolution process of forecasting system(s) are: Determining forecasting needs, Identity business drivers, develop channels of communications, Determine a system for forecasting and reporting and Constant improvement. (Wall,2002). “The forecasting need for this brand would require them to forecast the information by using multiple forecasting systems (MFS), the reason to this is because they had different data that needed to be evaluated individually by various models, then, later on, the models would be combined into one model for better accuracy. To create the models, they first had to pinpoint the means in which revenues are generated for the store; this included On-premises and off-premise channels. The on-premises channel would consist of areas such as retail stores, fundraising events, and the off-premises channel would include off site’s distribution centers, this included retail stores and convenience stores. Due to the abundance of data and information, it was decided to focus on three key driver (s):
· New Stores- they evaluated data that included predictions on the amount of business that the new stores would receive and when each store would open.
· New Off-Premises customers- The off-premise(s) location would include stores with high foot traffic, the greater the foot traffic, the higher the chances of customers purchasing the product. To evaluate such, they would need to know when their business began with the off-premise location to monitor the progress.
· Seasonality- They assessed seasonality because for each different channel-specific pattern can vary, especially between their off-premise and on-premises locations. Specific promotions and seasons can change the outcome or information obtained from channel to channel.
The information obtained came from monthly meetings between management all over a specific area or region. The reason that they decided to have these meetings is to keep constant communication between management and report on any changes. By keeping the flow of communication, they can bounce off ideas and information that can help with the forecasting procedure to obtain accurate results. Not only this but constantly exchanging information brings constant improvement to their brand, if there is a specific aspect of the forecast that is not working, they can communicate such and make any changes necessary. By implementing such system, they have a way to continually keep the flow of communication livid especially since they established a hub, this hub stores all information from different stores to use. Such information can be used to predict future values that can help them with a specified objective.
Another well-known brand that also uses forecasting to make management decisions is Brake Parts Inc, a leading private company that acts as a manufacturing and distribution facility. With nine manufacturing plants and seven distribution centers Brake parts, Inc has become a global secondary market for automotive parts. To account for all the different car parts Brake Parts, Inc has developed Multiple Forecasting systems (MFS). This system is used to keep everything accounted for and to prevent revenue loss from stockout periods, and it works in combination with another system called Management information system (MIS). MIS is a program that keeps track of any changes in their inventory; this program creates a file of such data, and then sends the data to the MFS for evaluation and forecast creation. The reason in which this company created MFS and MIS is because their company requires a forecast that is adaptable to different information systems. BPI has anywhere for 110,000- 135,000 Stock keeping unit (SKU), a forecast is generated for each unit.
The Multiple Forecasting System (MFS) created for this organization uses three essential methods of forecasting: Regression, Time series, and Judgmental. For the regression and time-series method, their system intertwines the data sent by their Management information system to generate a forecast, the reason why they do this is because regression model has specific information that the time series model does not provide and vice versa. Based on the article “Multiple Forecasting systems at Brakes Parts, INC their system uses up to nineteen-time series techniques. Once they have all nineteen time-series forecast their MFS system will evaluate the accuracy of each. The way that the system evaluates the forecast is by finding either the mean absolute error (MAE), Mean error (ME), Root mean squared error (RMSE), etc.· The forecast with the lowest error is the one that is selected, then the rest are discarded for the month. Once a forecast is selected management will meet and discuss the current forecast and also discuss the progress of any previous forecast implemented in the past. This part of the process is done to get personal opinions from the managers, who are in the industry and can provide valuable input.
Example three forecasting in the medical field:
Previously we have discussed how forecasting can be an essential asset in a retail organization that sells certain parts or goods. But forecasting is not limited, it can be used in any part of the business whether it is in the government, in a retail store and even in the medical field. When working on weekly homework, I came across a story in chapter 5, and this story was about how forecasting has helped hospitals in New Jersey successfully allocate their resources to budget for the future demand of Registered nurses. When working in a hospital, there has to be proper planning to ensure safety and patient satisfaction by properly allocating their resources. There has to be the right patient to Nurse ratio to ensure that each patient is receiving proper care. The article on page 245 “The Demand For nurses” spoke how Hospitals in NJ used forecasting to predict the demand for nurses using a multiple time series regression model. When establishing a forecast, there has to be an objective, a specific target to obtain better results. But with so much information collected from the hospitals of NJ what type of information can be used to predict the demand for nurses in the future? The hospitals decided to find data based on specific variables that contributed to the impact of the registered nurse demand. It was decided to use variables as, number of patients with HIV/AIDS because they did have a significant effect on the need for registered nurses, Birthrate, surgery rate, etc. All of these variables somehow where connected to the demand for nurses, if Birthrate increased then there have to be more Registered Nurses available to deliver the baby, same with surgery rate if the number of surgeries increases there was to be a certain number of registered nurses to help with the operation. In such environment, it is hard to predict the future outcomes, but by using forecast to make future predictions based of facts and personal judgment the hospitals can ensure the proper allocation and availability of their Registered nurse staff.
At the end I have a positive outlook on Business forecasting, it can be used in any environment to predict a particular outcome. For Krispy Kreme establishing a forecasting system helped them reduce waste, for Brake parts, INC it cut stockout periods, and for the hospitals in NJ, it helped them budget to create the right nurse to patient ratio. If you are looking to implement a forecasting system in your organization there has to be a clear objective, you need to know what you want and what sort of outcome you are expecting. If data is just being thrown around without knowing what you are looking for, information can be convoluting and time-consuming. At first, I had no idea of what forecasting was and what it can achieve when a specific procedure is followed, but after studying such process, this would be a procedure I would establish if I had my own business. When I was young, my father had a company, and one of the primary reasons that it went out of business is because he was losing money by not planning accordingly or making uninformed decisions. If he was aware of this process, I firmly believe that he could’ve made better-informed decisions based on facts and not solely on personal judgment.