Supply Chain Network Design

An imaginary eCommerce firm, Begpacker Inc. produces a series of 3D pop-up paper art (PopCards), becoming very popular in some Southeast Asia countries, notably Malaysia, Singapore, Thailand and Vietnam.

Its sales have been expanding rapidly. Their home-based production lines in Kuala Lumpur, which ships directly to customers, is now out of capacity. The CEO decided to have at least one proper facility for production and distribution. He is trying to decide the number and the locations of the new facilities to better meet projected demand over the next whole year, while minimizing the transportation costs.

Does this problem familiar to you? You can get your answers with Supply Chain Network Design.

What is Supply Chain Network Design?

Supply Chain Network Design is about determining the optimal number, size and locations of facilities (suppliers, plants, warehouses, … etc.) and the best flow of products between them.

Supply Chain Network Design helps you to determine:

The number of facilities.
The size / capacity / flexibility of the facilities.
The locations of facilities.
The territories for each facility.
The product should be made.
The flow of the products through the supply chain, from the source to the final point of consumption.

By implementing the recommended changes from your optimized Supply Chain Network Design, you’re looking at 5-15%, fairly large cost savings and service improvement from the original state.

How does Supply Chain Network Design work?

Supply Chain Network Design utilizes powerful optimization engine (like IBM ILOG CPLEX Optimization) and mathematical models or techniques to analyze a large amount of possible scenarios to provide you the actionable decisions.

As the result, you can reduce your costs (logistics, inventory, labor, utilities, taxes, …), your risks (fire, flood, natural disaster, strike, legal, political, …) and carbon emissions. As well as incresing your service level (customer satisfication) and flexibility of your business.

Oversimplified Case Study: Begpacker Inc.

The real world supply chain can be very complex. Consider the problem Begpacker Inc faced in the beginning: How can the CEO minimize his transportation cost while providing high service by meeting the customers’ demand? It is the most basic Supply Chain Network Design: deciding the number and locations of facilities (here plant-attached warehouse), based on the distance and demand.

City	Demand (unit)
Kuala Lumpur, MY	187,000
Penang, MY	35,000
Singapore, SG	190,000
Bangkok, TH	172,300
Chiang Mai, TH	57,200
Ho Chi Minh, VN	145,000
Nha Trang, VN	30,600
Danang, VN	61,000
Hanoi, VN	123,000

1. Add the cities into Google Map through the Search box.

2. Edit the demand for each city.

3. Try with the scenario of One facility (Plant-attached Warehouse).

For only one plant-attached warehouse, he found that Bangkok is a better place, which is more center than Kuala Lumpur, to fulfil the demand with less distance traveled (weighted average distance), hence the lower transportation costs.

4. More scenarios with multiple facilities

His goal is to determine how many facilities that he need to open, so he can just run multiple scenarios with the different number of facilities.

2 Facilities

The first facility will be staying in Bangkok, with the new one is opened in Kuala Lumpur.

3 Facilities

The third one should be opened in Hanoi. Interestingly, the demand of Nha Trang and Ho Chi Minh (Vietnam) will be fulfilled by Bangkok (Thailand). We did not consider the cross boarder cost for this case study.

4 Facilities

The forth facility should be opened in Ho Chi Minh, instead of Singapore with higher demand.

5 Facilities

The fifth facility should be opened in Singapore.

5. Comparing the Weighted Average Distance

Our objective is to minimize the costs based on demand, which is the weighted average distance. We can construct a table for each scenario.

Scenario (# Facility)	Weighted Average Distance (km*unit)	Improvement (%)	Differentiation (%)
0	1,382,776,200	0	20.19
1	1,103,537,500	20.19	42.26
2	519,147,500	62.46	13.48
3	332,804,500	75.93	10.87
4	182,524,300	86.80	5.07
5	112,377,400	91.87

Generally, the more facilities, the less distance (our objective), until we have one facility for each city.

For Scenario 0, the CEO will keep the original Kuala Lumpur as the only facility, so we cannot see any improvement. However, there is a 20.19% reduction in weighted distance, by just moving the only one facility from Kuala Lumpur to Bangkok.

After the 2nd facility, we can observe there are diminishing returns by adding more facilities. When we go from 1 facility to 2 facilities, the distance decreases by 42.26%! However, when we go from 2 facilities to 3 facilities, the distance decreases only by 13.48%.

In real world, by adding new facilities, will increase the cost of opening and operating the facility and other costs like cross boarder taxes, raw materials, labor, utilities, … etc. Although we’re unabled to quantify these costs in this case study, we knows that going from 1 facility to 2 facilities, will have the best improvement. After that, the subsequent smaller improvements are not worth the extra cost of additional facilities.

This case study seems to be too simplistic, i.e. it is only minimizing the weighted distance. However, this simple model can solve many supply chain network problems. Distance and transportation costs are highly correlated, i.e. the farther you have to go, the more expensive. This is a good approximation for minimizing the transportation costs. By having facilities as close to the customers as possible, indirectly providing better service levels.

If you’re interested to learn more how to model your real world supply chain and get great benefits from that, feel free to talk to us.

Reference:

Supply Chain Network Design Intro: Southeast Asia eCommerce Case