Question Michelle Prud\'homme is the plant manager for Personal Care Grooming, w

Question

Question

Michelle Prud'homme is the plant manager for Personal Care Grooming, which makes a variety of personal-care products such as combs, brushes, and razors. The company produces over 1,000 products, each of which is maintained as in-house inventory. The inventory performanceon these items in aggregate is fairly good with a 98 percent in-stock rate, but at the individual product level, there are products with from 0 to 385 days of inventory. Michelle does not have a good understanding of how the company forecasts demand for products because it is coded into the company's ERP system, and there is no documentation detailing how forecasts are calculated.

In order to assess how well the system is forecasting, she has selected three representative products. One of the products has a seasonal demand pattern, one of the prducts has a strong trend component, and one of the products has neither a seasonal nor a trend component. She has asked you to recommen one or more forecasting methods for each product and to evaluate the relative accuracy of the forecast compared to the existing method. As an additional test of your abilities, Michelle has not told you which product has which demand components. Utilize the data in the table to address the following issues:

1. Recommend a forecasting method for each product based on a review of the data. Calculate forecasts for each month in the table, plus forecasts for July and August of the current year.

2. Evaluate the performance of your recommended forecasting method using common measures of forecast error and compare it to the performance  of the current forecasting method.

3. Discuss the conditions and factors that might lead you to try other forecasting methods, such as qualitative or casual models. What types of data might be used ans an independent variable for a company with products like those of Personal Care Grooming?

B1300 SALES B1300 FORECAST A1000 SALES A100 FORECAST C190 SALES C190 FORECAST Jan 07 2000                6,532.1                   6,413                   1,683                   1,665 Feb 07 2120                        2,000                7,104.5                   7,378                   1,692                   1,683 Mar-07 1950                     2,063.2                7,096.7                   7,076                   1,586                   1,687 Apr 07 1910                 2,049.648                9,995.1                 10,043                   1,569                   1,654 May-07 1800                 2,019.783              10,740.7                 10,710                   1,960                   1,616 Jun 07 2100                   1,952.69              11,699.1                 11,590                   1,840                   1,705 Jul 07 1850                 2,004.564              12,505.7                 11,858                   1,823                   1,790 Aug 07 1950                 1,956.602              13,772.8                 13,566                   1,720                   1,874 Sep 07 1780                 1,952.632              11,124.6                 10,370                   1,681                   1,794 Oct-07 1650                 1,888.495              10,106.3                   9,680                   1,535                   1,741 Nov 07 1825                 1,790.290                5,992.2                   5,712                   1,538                   1,645 Dec-07 1500                 1,776.129                5,718.6                   5,612                   1,737                   1,585

Explanation / Answer

The first step would be to plot the data in excel (or any other plotter) and get the seasonality and trend components.

Further, separating out the graphs of A ( B1300 model) and C (C-190 model)

we see that, C-190 model has neither trend nor seasonality component
A-100 model has a strong seasonality component as it peaks in Jul and Aug
B-1300 model has a visible downward trend

Following models can be used:

Model Type

Most Suited Data Types

Forecast Horizon

Shelf Life of Model

Exponential Smoothing

No Trend, Varying Levels

Short

Short

Holt's Method

Varying Trends, Varying Levels, No Seasonality

Short

Short

Winter's Method

Varying Trends, Varying Levels and Seasonality

Short
to Medium

Medium

ARIMA

Varying Trends, Varying Levels,
Seasonality

Short
to Medium

Long

Now,

For, C-190 model we will use exponential smoothing method using alpha value as 0.1, 0.2 and 0.5

We shall use holt's method to determine to forecast the trend in the B-1300 sales

For the last forecasting we shall use Winter's method. However, we need at least one more year of data for that.

Model Type

Most Suited Data Types

Forecast Horizon

Shelf Life of Model

Exponential Smoothing

No Trend, Varying Levels

Short

Short

Holt's Method

Varying Trends, Varying Levels, No Seasonality

Short

Short

Winter's Method

Varying Trends, Varying Levels and Seasonality

Short
to Medium

Medium

ARIMA

Varying Trends, Varying Levels,
Seasonality

Short
to Medium

Long

Related Questions

Navigate

• Browse (All)
• Browse Q
• Subjects

---

---

Integrity-first tutoring: explanations and feedback only — we do not complete graded work. Learn more.