Introduction: A company is looking to open a new factory location (or close an old one) and is looking for your assistance in determining a location. How do you go about selecting where to open a new plant (or which old plant to close)?
Salience: There are many factors which are important in making this decision. For instance, how much capacity is required after the proposed changes? What is the distribution network needs based on geography? What is the cost of the factors of production (land, labour, capital) associated with different locations?
Causality: With the goal of optimal operations to achieve maximum profitability, each of these factors will have a different effect on how you make your decision. In closing an old plant, you will have to do a cost / benefit analysis of each plant and determine which one makes the most sense to shut down. The following framework can be adapted to better understand the closing of one factory to the entire manufacturing load and network.
In the scenario of opening a new plant, you technically have more flexibility in terms of which locations where you want to open (including even outsourcing capacity from others) so we can start to build a framework about how to decide what consitutes an optimal solution.
Architecture: There are many factors to consider in a holistic approach.
- Geographic capacity demand.
- Distribution of products.
- Local labour, material and transportation costs.
- Resource availability.
Geographic Center of Gravity First, let's simplify the model by assuming, cateris paribus, that the only thing that matters is geography. In this case, you can make an easy decision by taking mapping potential factories by using a
center of gravity formula. "Gravity" in this case is capacity demand. Also, factories currently in operation would serve as negative "gravity". This is because they are already servicing demand in the area. The resulting "center of gravity" would be a reflection of an area with the highest capacity demand.
R = Σ [Vi x Qi] / n
Where:
- n is the number of current factories in operation
- R is the optimal vector of your new factory location
- Vi is the vector describing the locations of your relevant capacity factors.
- Qi is a weighting applied to the relative capacity impact of each location (a positive value implies a customer demand, a negative value implies a factory capacity supplied). This factor can also be scaled for other factors accordingly.
- i is a counter variable iterating from 1 to n (encompassing all elements affecting capacity)
This formula assumes that each factory has identical weight in terms of capacity, costs etc. However, instead of a straight forward calculation of an average, each factory can be weighted with these additional factors to provide a more reasonable measure. For instance, each factory can be weighted with it's relative capacity.
Now, what if you only have a limited number of possibilities because of such factors as labour and resources are limited to big city areas etc? You need to match the profiles of your possible solutions to your "optimal" solution. However, in looking at your optimal solution, perhaps it will provide you with a potential solution that you had not previously considered (locating in a different town for instance).
Resolution: Although this is a very reasonable methodology, it only provides a mechanical answer based on the inputs provided and requires the analysts to accurately gauge the weight and importance of each individual factor. There may be many other influences such as political pressure to locate in a particular area. However, it acts as a logical framework for identifying the value of different locations while considering the broadest and more relevent factors relating to the capacity management decision.
[Case Study] A consulting company has 5 equally skilled consultants in the same field. 2 are in New York, one lives in Boston, MA and one in Philadelphia, PA. Their business is as follows is divided geographically as follows:
- 20% Philadelphia
- 50% New York
- 30% in Boston
Assume that each consultant is equally effective and the work is divided evenly. Also, the last consultant is more flexible to travel (but all consultants generally want to travel as little as possible), where should the last consultant reside?
Using the formula above, what is the optimal location for the last consultant to reside?
[
Answer: Hartford, Connecticut. Reasoning: Each consultant reflects 20% of the work load. This means that the consultant in Philadelphia can deal with the work load there. Two of the NY consultants reduce NY's capacity deficiency to 10% as does the consultant in Boston. Another way to look at the solution is that the only work left for this last consultant is equally split between New York and Boston.
The so in calculating the center of gravity, we learn that the optimal location solution is equidistant from New York and Boston (Hartford) - Note that Hartford was not a suggested location, but came up in the investigation.
Also note that the assumptions were just for simplicity in illustrating the solution, but the differences of the contributions, demands and travel costs of each individual component can be mathematically weighted against the whole - Philadelphia has more work demand than Boston, Senior Consultants do more work, costs for junior consultants is cheaper etc.]
This is a great structure to apply to make team members more aware of the current context for decision making and setting direction.
