At the beginning of 2018, Mark Columbus was appointed the new Head of Supply Chain at Supercoffee, a California coffee pod producer both for coffee shops (e.g., Starbucks) and households (e.g., Keurig machine).

His main objective was to develop more the global outreach of the company. Currently, Supercoffee distributes its products mainly in the local markets (United States and Canada), and Columbus objectives are to start exporting at least in some countries in Central and South America.

Supercoffee in-house production consists of 158 products, which can be grouped into five families: Strong, Extra, Aroma, Napoli, and America. Each family has a specific blend, roasting temperature, processing procedure (in beans or ground), labels for domestic and international markets, and they can be sold in multiple formats (lots of pods of 2.5 kg for coffee shops, 0.25 and 0.15 kg for households)

After an initial growth in 2010-2012, demand for coffee pods has overall stabilized in the final market, and companies like Supercoffee work with coffee machines producer and distributor to produce the right mix and quantity of coffee to satisfy customer request, as they are able to issue a forecast of their future orders for at least 3 months. This way, demand trends are usually quite well characterized for different types of products, and this gives also the possibility to companies to introduce innovations in the flavor and/or packaging, considering the long lifecycle characterizing these products.

More critical, instead, is the management of the upstream part of the supply chain, for three reasons. First, high-quality coffee beans are produced only in limited regions world-wide and, has experienced some shortage in the last years due to environmental (i.e., weather) and labor (i.e., strikes) factors. Second, pod packaging needs to be source by respecting specific requirements, considering the strong attention to sustainability aspects related to their disposal. For this, although Supercoffee tries to have a consolidated set of suppliers, frequent rotation and introduction of new partners is necessary. Last, production technologies for bean processing, blending, roasting, flavoring, and packaging are in constant evolution and require companies to invest (if possible) in frequent manufacturing process renovations.

Distribution network design

Currently, Supercoffee has one central warehouse, located in Reno (Nevada).

From this warehouse, coffee pods are sent by road and/or rail to several small and big retailers all over the United States and Canada.

Once Supercoffee marketing will have defined a clear strategy for entering foreign markets in South America, the supply chain must be ready to support these efforts. For these reasons, Columbus was thinking about a new structure of the distribution network.

Instead of reaching directly the industrial customers through the central warehouse, Columbus would like to use local warehouses (owned and executed by a logistic service provider, but managed and supervised by Supercoffee) in each of the final market (United States, Canada, and other Central and South American countries), in order to make the transportation from the central warehouse more efficient, but also leave open doors to possible personalization of variety according to the characteristic of the demand in each of these markets.

Inventory management policies

Considering the characteristics of the Supercofee supply chain, managing inventory of final products should not give the company too much of concern.

Instead, in the last two years, Supercofee production system struggled a lot to provide the central warehouse the right quantity at the right time, in order to satisfy retailers demand. The central warehouse does not hold safety stock, and several times, Supercoffee ended up sending urgent deliveries without saturating the trucks or the rail cars, thus ending up spending more logistics costs than planned.

Last year, Supercoffee had $ 151,081 in finished product inventory and spent $ 38,901 total inventory management costs (including both carrying costs and ordering costs), and $ 17,863 in transportation costs.

In order to reduce this problem, Columbus wants also to review the inventory management policy for finished products to be held in the central warehouse. In particular, in order to contain costs, he wants to implement an economic order quantity model with fixed reorder.

To quantify the parameter of the model, he has collected several pieces of information.

First of all, he collected the following data about production and transportation costs for the different product families.

Table 1. Production and transportation cost data of Supercoffee products ($ per unit, where 1 unit = 1 lot; each lot is made by 100 sleeves; each sleeve includes 12 pods)

	Strong	Extra	Aroma	Napoli	America
Production costs
Coffee blend	244.20	245.60	308.90	284.20	261.20
Materials for packing	12.70	12.70	12.70	12.70	12.70
Direct labor	1.00	1.00	1.00	1.00	1.00
Duties	9.77	9.83	12.75	11.10	10.38
Taxes	7.33	7.37	9.35	8.40	7.82
Other general variable costs	5.00	5.00	5.00	5.00	5.00
Other general fixed costs	13.10	13.10	13.10	13.10	13.10
Transportation costs
	1.59	1.64	2.2	2.05	1.82

Flag this Question

The annual inventory holding rate is the same for all the families, and it is composed of mainly financial opportunity cost of capital, equal to 9%, and defined according to the interest rate on loan received by SuperCoffee. The other costs (obsolescence, storage, insurance, etc.) associated to inventory holding are instead quite low, and amount to 2.5% all-in.

In addition to this, Columbus collected data about the reorder cost of each family (including both the set-up costs to switch from one family to the other and the average costs of administrative staff to issue the order for filling a lot from the distribution center):

• Strong: 278.3 $/order
• Extra: 278.3 $/order
• Aroma: 342.5 $/order
• Napoli: 303.1 $/order
• America: 292.4 $/order

Finally, Columbus collected the data about the forecasted demand for the next year, as well as the past values of replenishment lead time from the production plant (note: this value is equal for all the families of product).

Table 2. Summary of monthly forecasted orders coming from retailers (number of units aka lots)

	Strong	Extra	Aroma	Napoli	America
Jan	114	42	39	654	22
Feb	128	51	79	163	10
Mar	136	52	82	180	34
Apr	233	74	151	198	44
May	219	157	66	183	26
June	284	150	127	217	33
July	343	257	96	207	35
Aug	368	179	85	186	51
Sep	230	83	61	171	16
Oct	162	72	67	205	15
Nov	246	89	103	266	26
Dec	252	181	131	257	43

Table 3. Value of the replenishment lead times from the production plant to the central warehouse during the last year (values in month)

	Replenishment lead time
Jan	0.33 month
Feb	0.54 month
Mar	0.37 month
Apr	0.40 month
May	0.47 month
June	0.50 month
July	0.43 month
Aug	0.53 month
Sep	0.33 month
Oct	0.47 month
Nov	0.50 month
Dec	0.40 month

Question

Columbus would like to change the inventory management policy, but he doesnt think that there is a need for Supercoffee to introduce safety stock for any of the families of coffee. We can forecast and manage the demand pretty well for all the families, and our plants are very reliable for what concerns the replenishment lead time. I am pretty confident we are able to give our customers a level of service of 90-92%, he says.

Do you agree with Columbus that safety stocks are not needed in the Supercoffee central warehouse? Carefully elaborate on your answer.

Suggestions: This is more of a reasoning and conceptual questions. Please note that it is not required to calculate the safety stock in this question. You could, if you think it is useful to support your thesis, but you can answer this question with no safety stock calculation. To do that, the meaning and role of safety stock must be clear, and also you need to be sure you understand from which variables the safety stock value depends on. If you are able to identify this, it is pretty clear from the number provided if Columbus is right or wrong. Of course, the fact that you are not asked to calculate safety stock doesnt mean you dont have to use numbers! To answer this question, you will need numbers at some point. So, again, if you dont see the need to use any values you already calculate or new onesprobably you are not on the right path.