case study Sport Obermeyer 郭瑞祥教授 【本著作除另有註明外，採取創用CC「姓名標示－非商業性－相同方式分享」台灣3.0版授權釋出】 1 美國哈佛大學校徽(wikipedia)

Outline Part I:Sport Obermeyer Part II: Student present 2

What are the constraints of the supply chain? 3

Ans:) 4

What is the typical time line at Obermeyer? Manufacturing capacity 9/92 10/92 11/92 12/92 1/93 2/93 3/93 4/93 5/93 6/93 7/93 8/93 1st production order 50% 2nd production order 50% Deliver non-reactive capacity 2nd order information Material LT reactive capacity production LT

Which products are safer to produce early 6

Ans:) 7

Part I:Sport Obermeyer Part II: Student present Outline Part I:Sport Obermeyer Part II: Student present 8

第一題 決策目標： 缺貨與減價成本最小化。 限制條件： 1)「沒有最小訂單量」 2) 總量趨近10,000個 第一次訂單_a)無最小訂單量 2) 總量趨近10,000個 第一次訂單_a)無最小訂單量 9

假設：1)平均分散缺貨(stockout)的風險。 2)忽略價差。 常態分配下，令每項商品的顯著水準α相等。 亦即每樣商品缺貨機率(1-α)皆相等。 ˙訂貨量 Ｘi＝μi－Ｚασi 第一次訂單_a)無最小訂單量 10

∵ΣXi＝20,000－9,428Ｚα→10,000 ∴Ｚα≒1.06，α＝0.1445，存貨的機率為14.45%。 樣式 預測訂購量 兩倍標準差 初期訂購數量 初期訂購量 Gail 1,017 388 X1＝1,017－388Ｚα 606 Isis 1,042 646 X2＝1,042－646Ｚα 357 Entice 1,358 496 X3＝1,358－496Ｚα 832 Assault 2,525 680 X4＝2,525－680Ｚα 1,804 Teri 1,100 762 X5＝1,100－762Ｚα 292 Electra 2,150 807 X6＝2,150－807Ｚα 1,295 Stephanie 1,113 1,048 X7＝1,113－1,048Ｚα 2 Seduced 4,017 X8＝4,017－1,113Ｚα 2,837 Anita 3,296 2,094 X9＝3,296－2,094Ｚα 1,076 Daphne 2,383 1,394 X10＝2,383－1,394Ｚα 905 總計 20,000 9,428 約10,000左右 10,006 ∵ΣXi＝20,000－9,428Ｚα→10,000　 ∴Ｚα≒1.06，α＝0.1445，存貨的機率為14.45%。 第一次訂單_a)無最小訂單量 11

第二題 決策目標： 缺貨與減價成本最小化。 限制條件： 1)「最小訂單量為600 」 2) 總量趨近10,000個 2) 總量趨近10,000個 第一次訂單_b)有最小訂單量 12

假設： Type 1 parka (μi > 1200 )， safety factor = (μi – 1200)/σi 高風險 難預測 Type１ 左尾檢定 safety factor越大→越能確定「訂貨」 訂貨量：Ｘi＝μi－Ｚασi Type２ 右尾檢定 safety factor越大→越能確定「不需訂貨」 訂貨量：600 (如需訂貨) Type３ 將Type３留到最後做調整 訂貨量：600或 　　　　Max（600，Ｘi＝μi－Zασi） 低風險 易預測 第一次訂單_b)有最小訂單量 13

樣式 總訂購量 (測) 兩倍標準差 Safety Factor Ｚ (X=600) 初期訂購數量 Seduced 4,017 1,113 2.53 Ｘ1＝μi－Ｚασi Assault 2,525 680 1.95 Ｘ2＝μi－Ｚασi Electra 2,150 807 1.18 Ｘ3＝μi－Ｚασi Anita 3,296 2,094 1.00 Ｘ4＝μi－Ｚασi Daphne 2,383 1,394 0.85 Ｘ5＝μi－Ｚασi Entice 1,358 496 0.32 Ｘ6＝μi－Ｚασi Gail 1,017 388 0.00 1.08 Ｘ7＝？ Isis 1,042 646 0.68 Ｘ8＝？ Teri 1,100 762 0.66 Ｘ9＝？ Stephanie 1,048 0.49 Ｘ10＝？ 總計 20,001 約10,000左右 第一次訂單_b)有最小訂單量 14

用試誤法調整Type 3 ： Ｘi＝ μi－Ｚασi ，( i＝1、2…6 ) Ｙi＝Max（600, μi－Ｚασi） 樣式 方案一 方案二 方案三 方案四 方案五 X1~X6 μi－Ｚασi X7 600 Ｙ7 X8 Ｙ8 X9 Ｙ9 X10 Ｙ10 15

樣式 總訂購量 (測) 兩倍標準差 Safety Factor Ｚ (X=600) 初期訂購數量 Seduced 4,017 1,113 2.53 Ｘ1＝μi－Ｚασi Assault 2,525 680 1.95 Ｘ2＝μi－Ｚασi Electra 2,150 807 1.18 Ｘ3＝μi－Ｚασi Anita 3,296 2,094 1.00 Ｘ4＝μi－Ｚασi Daphne 2,383 1,394 0.85 Ｘ5＝μi－Ｚασi Entice 1,358 496 0.32 Ｘ6＝μi－Ｚασi Gail 1,017 388 0.00 1.08 Ｘ7＝？ Isis 1,042 646 0.68 Ｘ8＝？ Teri 1,100 762 0.66 Ｘ9＝？ Stephanie 1,048 0.49 Ｘ10＝？ 總計 20,001 約10,000左右 第一次訂單_b)有最小訂單量 16

用試誤法調整Type 3 ： Ｘi＝ μi－Ｚασi ，( i＝1、2…6 ) Ｙi＝Max（600, μi－Ｚασi） 樣式 方案一 方案二 方案三 方案四 方案五 X1~X6 μi－Ｚασi X7 600 Ｙ7 X8 Ｙ8 X9 Ｙ9 X10 Ｙ10 17

僅［方案一］成立，Ｚα＝1.234 樣式 總訂購量(預測) 兩倍標準差 初期訂購數量 Seduced 4,017 1,113 X1＝4,017－1,113Ｚα 2,644 Assault 2,525 680 X2＝2,525－680Ｚα 1,686 Electra 2,150 807 X3＝2,150－807Ｚα 1,154 Anita 3,296 2,094 X4＝3,296－2,094Ｚα 712 Daphne 2,383 1,394 X5＝2,383－1,394Ｚα 663 Entice 1,358 496 X6＝1,358－496Ｚα 746 Gail 1,017 388 X7＝600 600 Isis 1,042 646 X8＝600 Teri 1,100 762 X9＝600 Stephanie 1,048 X10＝600 總計 20,001 10,005 第一次訂單_b)有最小訂單量 18

講 評 沒有限制最小訂購量 公司彈性增加：對於那些風險性較高的樣式Obermeyer可以在第一階段的生產上生產較少產量 在資訊揭露後：進行第二次的reactive planning，可減少存貨，讓生產與市場需求配合。 19

How do you plan the initial production without considering minimum order size? Assume normal distribution for each style’s forecast Produce up to the same demand percentile for each style Stylei = μi – kσi (vary k so that total is around 10000) Style μi σi μi – 1.06σi Seduced 4017 1113 2837 Assault 2525 680 1804 Electra 2150 807 1295 Anita 3296 2094 1076 Daphne 2383 1394 905 Entice 1358 496 832 Gail 1017 388 606 Isis 1042 646 357 Teri 1100 762 292 Stephanie 1048 2 Total 20001 10008

safety factor-I 假設： Type 1 parka (μi > 1200 )， safety factor = (μi – 1200)/σi Type 2 parka (μi < 600 )， 　 safety factor = (600–μi )/σi Type 3 parka (600< μi < 1200 )， safety factor = 0 高風險 難預測 Type１ 左尾檢定 safety factor越大→越能確定「訂貨」 訂貨量：Ｘi＝μi－Ｚασi Type２ 右尾檢定 safety factor越大→越能確定「不需訂貨」 訂貨量：600 (如需訂貨) Type３ 將Type３留到最後做調整 訂貨量：600或 　　　　Max（600，Ｘi＝μi－Zασi） 低風險 易預測 第一次訂單_b)有最小訂單量 21

safety factor-II 22

safety factor-III 23

safety factor-IV 24

How do you plan the initial production considering minimum order size of 600? Large production minimum order size forces us to make either many parkas of a given style, or none. Thus our ability to finely tune our supply of each parka is reduced significantly by the minimum order size. Define safety factor: Type1 parka (μi > 1200), safety factor = (μi - 1200)/σi Type2 parka (μi < 600), safety factor = (600 - μi)/σi Type3 parka (600 <μi < 1200), safety factor = 0

The initial production plan Production Quantity = Max (600, μi – 600 – SFmin * σi) Style μi σi Safety Factor Production quantity Seduced 4017 1113 2.53 3062 Assault 2525 680 1.95 1708 Electra 2150 807 1.18 1293 Anita 3296 2094 1.00 2029 Daphne 2383 1394 0.85 1339 Entice 1358 496 0.32 600 Total 10031

The impact of minimum order size

The benefits of increasing reactive production capacity

How to increase reactive capacity? 29

Ans:) 30

How to increase reactive capacity? Increase reactive capacity only Develop a workforce willing to work longer hours during peak season (workers in Lo Village work nearly 50% more hours) Use subcontractor during peak season Increase total capacity Increase the total (year-round) capacity Manufacture complementary garments such as swim suits Reduce raw material and/or manufacturing lead times Short production lead time → more production units Raw material lead time is almost three months now Stock inventory of long-lead time materials Increase component commonality (e.g., zippers) Obtain market information earlier Offer discounts to early orders Choose early-order retailers form different segments and update its demand forecast accordingly

How to increase reactive capacity? a. Base Case Speculative Production Capacity Original Reactive Capacity Material LT New Info Prod’n LT Material LT Prod’n LT New Info Note: The shaded areas indicate the time at which production begins. Production is completed at an addition amount of time (the production lead time) has elapsed b. Increase Total Capacity Additional Reactive Capacity

How to increase reactive capacity? Additional Reactive Capacity c. Decrease Lead Times Material LT New Info Prod’n LT Material LT Prod’n LT New Info d. Obtain Informational Earlier Additional Reactive Capacity Prod’n LT Material LT New Info

What are the roles of Sport Obermeyer and Obersport? 34

What are the roles of Sport Obermeyer and Obersport? Supply chain Obersport Sport Obermeyer Market Obersport Sport Obermeyer Coordinate supply side market side Tackle demand Monitor fashion trend Monitor retailer’s willingness and abilities to pay Monitor the health of the US economy Sign contract with retailers Offer discounts Find component suppliers Build relationship with the suppliers Find qualified factories Monitor the quality and schedule of the produced products

What are the comparisons of operations in HK and China? Flexible hours Low High Labor cost Worker productivity Minimum order size 600 1200 Quality Challenges Wage rate high Low unemployment Less quality Training

版權聲明 1, 4, 7, 8, 29, 30, 34 37 頁碼 作品 授權條件 作者/來源 美國哈佛大學校徽。 本作品轉載自WIKIPEDIA (http://zh.wikipedia.org/wiki/%E5%93%88%E4%BD%9B%E5%A4%A7%E5%AD%A6 )，瀏覽日期2012/1/2，依據著作權法第46、50、52、65條合理使用。 37