In: Math
Jan Northcutt, owner of Northcutt Bikes, started business in 1995. She notices the quality of bikes she purchased for sale in her bike shop declining while the prices went up. She also found it more difficult to obtain the features she wanted on ordered bikes without waiting for months. Her frustration turned to a determination to build her own bikes to her particular customer specifications.
She began by buying all the necessary parts (frames, seats, tires, etc.) and assembling them in a rented garage using two helpers. As the word spread about her shop’s responsiveness to options, delivery, and quality, however, the individual customer base grew to include other bike shops in the area. As her business grew and demanded more of her attention, she soon found it necessary to sell the bike shop itself and concentrate on the production of bikes from a fairly large leased factory space.
As the business continued to grow, she backward integrated more and more processes into her operation, so that now she purchases less than 50% of the component value of the manufactured bikes. This not only improves her control of production quality but also helps her control the costs of production and makes the final product more cost attractive to her customers.
The Current Situation
Jan considers herself a hands-on manager and has typically used her intuition and her knowledge of the market to anticipate production needs. Since one of her founding principles was rapid and reliable delivery to customer specification, she felt she needed to begin production of the basic parts for each particular style of bike well in advance of demand. In that way she could have the basic frame, wheels, and standard accessories started in production prior to the recognition of actual demand, leaving only the optional add-ons to assemble once the order came in. Her turnaround time for an order of less than half the industry average is considered a major strategic advantage, and she feels it is vital for her to maintain or even improve on response time if she is to maintain her successful operation.
As the customer base have grown, however, the number of customers Jan knows personally has shrunk significantly as a percentage of the total customer base for Northcutt Bikes, and many of these new customers are expecting or even demanding very short response times, as that is what attracted them to Northcutt Bikes in the first place. This condition, in addition to the volatility of overall demand, has put a strain on capacity planning. She finds that at times there is a lot of idle time (adding significantly to costs), whereas at other times the demand exceeds capacity and hurts customer response time. The production facility has therefore turned to trying to project demand for certain models, and actually building a finished goods inventory of those models. This has not proven to be too satisfactory, as it has actually hurt costs and some response times. Reasons include the following:
- The finished goods inventory is often not the “right” inventory, meaning shortages for some goods and excessive inventory of others. This condition both hurts responsiveness and increases inventory costs.
- Often, to help maintain responsiveness, inventory is withdrawn from finished goods and reworked, adding to product cost.
- Reworking inventory uses valuable capacity for other customer orders, again resulting in poorer response times and/or increased costs due to expediting. Existing production orders and rework orders are both competing for vital equipment and resources during times of high demand, and scheduling has become a nightmare.
The inventory problem has grown to the point that additional storage space is needed, and that is a cost that Jan would like to avoid if possible.
Another problem that Jan faces is the volatility of demand for bikes. Since she is worried about unproductive idle time and yet does not wish to lay off her workers during times of low demand, she has allowed them to continue to work steadily and build finished goods. This makes the problem of building the “right” finished goods even more important, especially given the tight availability of storage space.
Past Demand
The following shows the monthly demand for one major product line: the standard 26-inch 10-speed street bike. Although it is only one of Jan’s products, it is representative of most of the major product lines currently being produced by Northcutt Bikes. If Jan can find a way to sue this data to more constructively understand her demand, she feels she can probably use the same methodologies to project demand for other major product families. Such knowledge can allow her, she feels, to plan more effectively and continue to be responsive while still controlling costs.
Actual Demand |
||||
Month |
2011 |
2012 |
2013 |
2014 |
January |
437 |
712 |
613 |
701 |
February |
605 |
732 |
984 |
1291 |
March |
722 |
829 |
812 |
1162 |
April |
893 |
992 |
1218 |
1088 |
May |
901 |
1148 |
1187 |
1497 |
June |
1311 |
1552 |
1430 |
1781 |
July |
1055 |
927 |
1392 |
1843 |
August |
975 |
1284 |
1481 |
839 |
September |
822 |
1118 |
940 |
1273 |
October |
893 |
737 |
994 |
912 |
November |
599 |
983 |
807 |
996 |
December |
608 |
872 |
527 |
792 |
1. Plot the data and describe what you see. What does it mean and how would you use the information from the plot to help you develop a forecast?
2. Use at least two different methodologies to develop as accurate a forecast as possible for the demand. Use each of those methods to project the next four months demand.
3. Which method from question 2 is “better”? How do you know that?
Answer - 1
If we plot the graph for past actual demand data. We observe that the demand increases January to June and it again decreases from June to December gradually.
Answer - 2
In this approach, the predictions of all future values are equal to the mean of the past data. This approach can be used with any sort of data where past data is available. In time series notation:
where is the past data.
Although the time series notation has been used here, the average approach can also be used for cross-sectional data (when we are predicting unobserved values; values that are not included in the data set). Then, the prediction for unobserved values is the average of the observed values.
Naïve forecasts are the most cost-effective forecasting model, and provide a benchmark against which more sophisticated models can be compared. This forecasting method is only suitable for time series data. Using the naïve approach, forecasts are produced that are equal to the last observed value. This method works quite well for economic and financial time series, which often have patterns that are difficult to reliably and accurately predict. If the time series is believed to have seasonality, the seasonal naïve approach may be more appropriate where the forecasts are equal to the value from last season. In time series notation:
Answer - 3
If we compare both the forecasting methods, Naive approach is better as it is most cost effective forecasting models which is need of this business scenario. The demand is fluctuating and Naive approach will help the planner in better inventory planning of finished goods and raw material.