Bio-diesel plant location decision.

Metlen, Scott ; Haines, Doug ; McAlexander, Amanda 等

CASE DESCRIPTION

This case addresses biodiesel production plant location considerations. The case is appropriate for undergraduate seniors (difficulty level: 4) in supply chain management, logistics, and/or general operations and marketing classes. Understanding the business issues presented is critical to firm success thus, to a student's success when they become involved in such decisions. The time a student must spend on this case for total understanding will vary depending on a student's base level of understanding, but most business students should be able to complete the case in four to six hours out of class and one hour of class discussion. The case is thirteen pages long, including references and appendices.

CASE SYNOPSIS

Bruce Nave had been using biodiesel in his own construction operation for over a year. With the advent of petroleum oil prices breaking seventy dollars per barrel, he saw an opportunity to start producing biodiesel on a commercial scale. Bruce knew that the success of his planned enterprise would depend in part on location, as each location would have different start up cost, cost of living, local laws, cost of doing business, availability and cost of inputs, and cost of shipping raw materials and finished product. Differences in these costs could quickly erode the slim contribution margins that commodity items generate. The case ends with Bruce wondering where he should locate his biodiesel production facility. The purpose of this case is to provide a decision scenario to students that will be managing supply chains, logistic functions of a firm, and/or are otherwise involved in strategic decisions relative to location of capacity.

INSTRUCTORS' NOTES

Recommendations for Teaching Approaches

With petroleum based diesel continuing to erode profits, Bruce Nave, president of Western Industrial Resources Corporation (WIRC), started using biodiesel in 2005 to help fuel the company's diesel engines. These engines provided power to 65 trucks and an assortment of welders, compressors, and earth moving equipment. Based on the successful use of biodiesel and the rising cost of petroleum diesel, he decided that producing biodiesel on a commercial scale was a viable business. Bruce decided to start a commercial bio-diesel operation named Wi BioFuels, Inc. Once Bruce made the decision to construct a bio-diesel production facility he had to decide where to locate that facility.

Location decisions significantly affect the profit margins and eventual success of a firm due to many variables. These variables may include availability and proximity of raw materials, appropriate labor, regulations and tax structure, proximity of customers, governmental incentives, proximity of competition, quality of life, proximity to information, cost and availability of utilities, real estate costs and availability, and construction costs (Stevenson, 2007).

Presented in the case that accompanies this teaching note is information that enables students to mimic the location decision making process that Bruce performed to make his decision. The case ends with Bruce wondering where he should locate the bio-diesel plant.

The purpose of this case is to facilitate discussion of location decisions and the impact that such decisions can have on firm profitability. Following are teaching objectives this case was designed to fulfill.

* To provide a situation that demonstrates the intractable nature of location decisions

* To provide information such that students can delineate significant factors that shape location decisions

* To provide the opportunity for students to model cost parameters of location factors to determine a quantitative location solution.

* To provide the opportunity to incorporate qualitative factors with the quantitative solution to determine the final location solution

Following are the corresponding learning objectives that students should be able to demonstrate.

* Explain and/or demonstrate the intractable nature of location decisions

* Delineate significant factors of a location scenario

* Analyze the quantitative and qualitative parameters of a location scenario

* Utilize the information from the data analyses to reach a location decision

* Communicate the location decision with all supporting evidence in a professional format

The material in this case is appropriate for upper division undergraduate or MBA level courses in Supply Chain Management, Logistics, Strategy, Channels and Distribution, and/or Operations Management that includes discussions dealing with plant location.

The information and situation detailed in the case reflects actual information encountered in an actual situation by actual people. The information and characteristics of the situation were obtained through interviews with the president of WIRC and Wi-BioFuels, Inc. and through secondary research concerning the possible locations, cost parameters, and the location decision literature.

Discussion of this case should incorporate a qualitative and quantitative location framework such as the frameworks presented by Vonderembse and White (2004) (Appendix 1). Students should carefully read the case to determine the important factors influencing the location decision. Then they should systematically perform a cost benefit analysis of each site. This analysis will produce a quantitative solution to the location decision that then needs to be tempered by the qualitative analysis that the students must also perform. The following are possible discussion questions.

1. What makes location decisions intractable?

2. What criteria should be used to determine the location of the facility?

3. What qualitative criteria should be used for this case and what weight should be given to these different criteria?

4. What tool/s should be used to inform the location decision and what is the outcome of using this/these tools?

5. Where should Bruce locate his plant and why?

Triangulation between several production and operations management books such as Stevenson (2007), Slack (2007), and Vonderembse (2004) can provide basic decision frameworks and a list of possible variables that affect location decisions. The framework and variables chosen, weight placed on each variable, and decision made by each student may and probably will differ. Following suggestions by Slack (2007), the following decision placed a high weight on the supply chain network possibilities associated with a given location. To determine levels of some qualitative variables for each site, students can utilize web sites and/or contact appropriate organizations such as port authorities. Following are possible answers to the discussion questions listed above.

1. What makes location decisions intractable?

Location decisions can be very intractable due to the different factors (see answer number 2) that affect the degree of current and future profitability that a given site can provide. These factors simultaneously affect different goals a company may be trying to achieve; some positively and some negatively. These effects may actually show that there are several locations equally desirable, even though for different reasons (Stevenson, 2007). For example, to minimize freight costs a company may want to locate their distribution warehouse in a central location relative to their weighted markets. An example of how different factors can offset each other is that a firm might find the ideal location relative to freight costs in an isolated area with no available infrastructure and/or cultural life. In trying to minimize total costs by locating in that place, the location may actually raise costs due to employee related qualitative issues such as general dissatisfaction resulting in high turnover, shirking, and/or sabotage. Therefore, it is possible that a quantitative analysis between location possibilities can be misleading and qualitative information needs to be used to differentiate between locations.

2. What factors/variables should decision makers use to determine the location of a facility?

Analyzing different locations and how well they will help the company reach their objectives is important because there generally is not going to be one simple perfect location, but there are usually several nearly equally acceptable options. As suggested by Stevenson (2007) in this case the availability and proximity of raw materials, appropriate labor, regulations and tax structure, proximity of customers and competition, governmental incentives, proximity of competition, quality of life, proximity to industry related information, cost and availability of utilities, real estate costs and availability, and construction costs are some different criteria that should be considered while going through the location decision process. Slack (2007) further explored supply chain effects that a location may have on firm results such as the network relationships within a supply chain.

Some examples presented by Stevenson (2007) include the importance of looking at the availability and proximity of raw materials, customers, and transportation costs. Locating in an area with available labor with the correct skill set will also help in recruiting human resources. If the correct labor is not available cost may be affected adversely, as the firm may have to pay overqualified people to do a job someone else can do for minimum wage or pay overqualified people minimum wage and have a discontented work force causing low productivity. Regulations and tax structure are going to have an effect on whether you can even have your plant in the area, how much extra taxes and fees will cost you, and if you can possibly receive incentives or be exempt from some fee structures.

In addition to factors that can be quantified, qualitative factors should also be explored when considering location. Stevenson (2007) suggests that education opportunity, shopping, recreation, transportation, religious worship options, available entertainment, quality of police, fire, and medical services, and size of a community make a community desirable for its workers. Satisfied workers can increase high quality production of a facility and these factors can affect employee satisfaction Stevenson (2007). Thus, these factors should also be analyzed for this case.

3. Relative to this case, what criteria (factors/variables) should be used for this case and what weight should be given to these different criteria?

Students should start with a complete list of factors important to location decisions developed through sources such as Slack et al (2007), Stevenson (2007), and Vonderembse and White (2004). Then, as they read through the case, they will find information to support key factors from their list of general factors found in their research. The weight that is placed on each factor should be first determined from the case; secondly, if the weight cannot be determined from the case, it should be determined and supported from each student's perspective. It should be noted that the answer supplied put the heaviest weight on the access to the farmers' network, as a production facility with no or relatively expensive raw material will quickly go out of business. The total annual cost of production should be calculated and compared across sites. However, the most expensive site may be selected because of qualitative factors. Factors that students should be able to see as important from reading the case and conducting research are presented below. Students should be encouraged to carefully reduce the list to only the more relevant factors to make the analysis more manageable.

Quantitative Market availability and cost of transportation Raw input availability and cost of transportation and contracting Variable costs, including permits, utilities, bonds, waste disposal, and property rent Local tax rates Qualitative Employee Satisfaction Real estate availability and cost Hospital availability * K-12 school rating University education availability Culture availability Airport availability Ethnic diversity * Local tax rates Plant functionality Labor availability Labor skill level * Proximity to a university with biodiesel experts Hospital availability * Chemical treatment center * Site availability * Ethnic diversity * Production growth possibilities * Feedstock growth possibilities * Local demand growth * Airport availability Local tax rates Supplier network to ensure feedstock supply

* not included in this comparison

A danger of having too many qualitative criteria is that the importance weight of each variable becomes meaningless. A good rule to use when selecting both qualitative and quantitative variables for the final analysis is to make sure the variables are relevant to the specific decision and actually vary across the sites. If there is no variation between prospective sites for a given variable, state that it does not and do not include it in the weighted analysis between sites. In addition, students should use the 80/20 rule to minimize their list of qualitative variables that are weighted and used to help make a location decision. A student would start with a complete list and then eliminate when possible, communicating their assumptions and reasons for keeping or eliminating the various variables. For this case, availability of health care, facility site availability, and growth possibilities with respect to facility and feedstock did not vary across sites. Labor skill level was also discounted, as 4/5s of the required work force was blue collar. In addition, ethnic diversity was not included, as, historically, it has not been a factor in firm success in the northwest. Although local demand growth differed, it did not vary by much and if the percent of biodiesel in a petroleum and biodiesel mix were increased beyond 20%, there would be ample increased demand at all sites to justify growth.

As stated before, scores and weights for each qualitative factor will vary by student. However, it is important that each student supports the scores and weights they assign. The method used to create Table 7 of Appendix 1 was to assess the standing of each site for a given variable and give the site with the best rating for a given factor a ten, then adjust the others down from there. Weights for the factors were determined by logic. The highest weights were given to factors that directly affected the functionality of the facility with lesser weights given to the employee satisfaction factors, as these variables indirectly affect the success of a facility. The factor with the highest weight was having an established network with local farmers because when there is no raw material, there is no chance of success. This decision was based on the Slack et. al. (2007) suggestion that a well developed supply chain is an important strategic advantage. Table 6 shows that Richland is the lowest cost location. However, information presented in Table 7 shows that Clarkston is the best site. The information in Table 6 shows that the costs of operating in Clarkston would be 16% greater than operating in Richland. Analysis shows that the major cause of the difference is the established supplier network. Bruce's long term connections in the Palouse area assure a sufficient number of farmers can be convinced to adjust their traditional rotation crops to supply sufficient oil seed feedstock for Wi-BioFuels' plant. To determine the actual importance of such a network, the cost of shipping all inputs for the feedstock from the Palouse area to Richland was included in the cost analysis for the Richland site. Under that scenario, the Richland site is nearly 20% more expensive than Clarkston. This information would suggest that Clarkston is the most favorable site.

Supply chain management literature such as that found in Slack et. al. (2007) emphasizes the importance of supplier networks. However, the case only briefly mentions Bruce's connection to such a network. Thus, an instructor of this case may want to emphasize this point during case discussion.

4. What tool/s should be used to inform the location decision and what is the outcome of using this/these tools?

A major tool that could be used for the cost variables is linear programming, but with only three locations to decide from it is not necessary. A simple table can be constructed in excel that shows factor costs and total costs for each location. An additional table can display the results of weighting the qualitative factors after first scoring each qualitative factor for each possible site (See Appendix 1). In the following analysis, a 10 point scale was used with 1 being low and 10 high to score the different qualitative variables.

5. Where should Bruce locate his plant and why?

According to the following analysis with the given weights, Bruce should build his plant in Clarkston at the Port of Wilma. The total costs per year are lower than for Portland, but 16% higher than Richland. However, the qualitative weighting is much higher for Clarkston than for Richland. The difference in costs only amounts to $169,942.81 per year, favoring Richland over Clarkston. However, the major factor influencing the difference in the qualitative assessment is access to the farmers' network that will ensure feedstock availability. Such access is critical to maintaining the costs as depicted. Thus, the recommendation is that Clarkston be the site of the new biodiesel production facility.

Appendix 1: Assessment Tables and Explanations

An estimate of total diesel usage for each area is displayed in Table 1. There was no acreage to grow oil crops listed in the case for the St. John area, so the only local demand would be for what is consumed in the city area. In addition, it was acknowledged in the case that the state of Washington had mandates of biodiesel usage, thus only Spokane and Seattle (both in Washington) are supplied by all areas in the table. The locale farm area for Clarkston includes both the Palouse and Camas Prairie areas, as both central locations of these areas are within 80 miles of Clarkston. As displayed in the table, both Clarkston and St. John facilities will have to ship fuel to the Richland area local users to sell all five million gallons of fuel.

Illustrated in Table 4 is the information that all areas have enough acreage to supply enough oil bearing crops to supply a five million capacity production facility.

Illustrated in Table 5 are the freight costs. The case and a quick look at an area map provides information to complete this table. The inbound freight from the Camas Prairie area with Grangeville the center to Clarkston is all truck; from there it can change to barge. Inbound freight from Rosalia to Clarkston can be by private railroad and from Moses Lake commercial railroad. Inbound to Richland from the Camas Prairie would be by truck to Clarkston and barge from there to Richland. Inbound to Richland from Rosalia and Moses Lake would be by train. Inbound to Portland would be a combination of truck and barge from the Camas Prairie and by train and barge from the other two areas. All outbound freight is by truck.

Information in Table 6 is derived from secondary research listed in Appendix 1 of the case and from Table 5 above.

Information displayed in Table 7 is from the case or Appendix 1 of the case. Heavier weights were placed on factors that directly affected facility functionality.

REFERENCES

Slack, Nigel, Stuart Chambers, & Robert Johnston. (2007). Chapter 6: Supply Network Design. Operations Management 4th E. Pearson Education, Inc. Retrieved on June 20, 2007, from http://wps.pearsoned.co.uk/ema_uk_he_slack_opsman_4/ 0,8757,1144898,00.html.

Stevenson, W. J. (2007). Operations Management (9th Ed.) Boston: McGraw-Hill.

Vonderembse, M. A. & White, G. (2004). Operations Management: Concepts, Methods, and Strategies. NJ: Wiley & Sons for Leyh Press LLC.

Scott Metlen, University of Idaho

Doug Haines, University of Idaho

Amanda McAlexander, University of Idaho

Table 1: Demand by Location For Purpose of Outbound Freight
Available Yearly Markets by gallon (7.6 lbs/gallon)

 Local Farm Area
 Spokane Seattle Clarkston

Diesel Use 1,370,000 * 20,000,000 * 18,980,000 **

Biodiesel 27,400 400,000 379,600
 demand at 2%
Biodiesel 274,000 4,000,000 3,796,000
 demand at 20%
Expected demand To Spokane To Seattle To Local Area ***
 by site
From Clarkston 27,400 400,000 3,796,000
From Richland 27,400 400,000 4,572,600
From St. John 27,400 400,000 131,200

 Local Farm Local City
 Area Richland area St. John

Diesel Use 73,000,000 ** 6,560,000 **

Biodiesel 1,460,000 131,200
 demand at 2%
Biodiesel 14,600,000 1,312,000
 demand at 20%
Expected demand To Richland Starting
 by site local area Production of
 Facility
From Clarkston 776,600 = 5,000,000
From Richland 0 = 5,000,000
From St. John 4,441,400 = 5,000,000

* numbers from case

** numbers from research

*** Reasoning to determine usage per area was derived from the
case where local farmers would be more willing to use a
higher percent of biodiesel in their engines than commercial
users. 'Local' usage based on acres times 7.3 gallons/acre/year

Table 2: Mileage Chart

 St. Johns/
Factor Clarkston Richland Portland

Miles Grangeville to: 74.33 211.21 418.06
Miles Rosalia to: 77.45 153.77 360.62
Miles Moses Lake to: 153.57 81.16 286.62
Miles Des Moines to: 1557.36 1647.2 1789.92
Miles Spokane to: 105.86 145.81 351.29
Miles Seattle to: 317.77 218.84 173.58
Miles St. John to: 344.16 226.47 7.63
Miles Clarkston to: 0 137.31 344.16
Miles Richland to: 137.31 0 226.47

Figures obtained through Research at Mapquest.com

Table 3: Production per Acre and Freight Costs Per Pound per Mile

Production Camas Palouse Columbia
per Acre Prairie Area Basin

 2880 1555 1500

 Private
Freight Costs Truck Train Train Ocean Barge

$/ton/mile 0.15 0.12 0.13 0.02 0.08

Table 4: Acres Needed and Available per Area

5 million gallons of biodiesel requires 5 million gallons of oil,
given all oil comes from canola

Given the following

 5,000,000 gallons capacity of facility
 38,000,000 pounds of oil at 7.6 lbs/gallon
 95,000,000 pounds of seed needed at 40% oil

Then how many acres of canola Camas Palouse Columbia
needed given production rates Prairie Region Basin

lbs per acre 2,880 1,555 1500
acres needed 32,986 61,093 63,333
Acres available by area given 150,000 500,000 2,500,000
 a four year rotation

Table 5: Cost of Freight (least cost method available for
each distance)

Inbound

 From/to Clarkston Richland St. John/Portland

Camas Prairie $529,601.25 $1,051,379.25 $1,837,409.25
Palouse $478,253.75 $876,489.00 $1,786,061.75
Columbia Basin $875,349.00 $462,612.00 $1,323,198.00
min/area $478,253.75 $462,612.00 $1,323,198.00

Outbound of Biodiesel from Production Facility to Locations
Identified in the Case

 Ship to
 Richland local
From/to Spokane Seattle Local Area Area

Clarkston $1,653.32 $72,451.56 $167,580.11 $60,781.92
Richland $2,277.26 $49,895.52 $211,533.96 $ -
St. John $5,486.45 $39,576.24 $5,982.72 $573,331.00

 Total in and out
 with inbound to
 Total Richland from
From/to Outbound Total Freight the Palouse area

Clarkston $302,466.91 $780,720.66 $780,720.66
Richland $263,706.74 $726,318.74 $1,140,195.74
St. John $624,376.41 $1,947,574.41 $1,947,574.41

Table 6: Yearly Site Variable Operating Costs (from
secondary research)

 Clarkston Richland St. John/Portland

Permits $75,000.00 $20,000.00 $100,000.00
Utilities $96,000.00 $60,000.00 $120,000.00
Bond $120,000.00 $80,000.00 $400,000.00
Waste Disposal $60,000.00 $24,000.00 $120,000.00
Rent of property $120,000.00 $180,000.00 $300,000.00
 Total $471,000.00 $364,000.00 $1,040,000.00


Total Costs Clarkston Richland

Freight Costs $786,238.15 $723,295.34
Variable Costs $471,000.00 $364,000.00
Total $1,257,238.15 $1,087,295.34

 Richland with
 St. John/ inbound from
Total Costs Portland the Palouse area

Freight Costs $1,791,552.22 $1,140,195.74
Variable Costs $1,040,000.00 $364,000.00
Total $2,831,552.22 $1,504,195.74

Clarkston--Richland = $169,942.81 When Clarkston/Richland 1
inbound freight is from the closest Cost Ratio
growing area

If inbound freight to Richland were Richland/Clarkston 1
from the Palouse area then: Cost Ratio
Richland--Clarkston = $246,957.59

Table 7: Qualitative Factors Analyzed

Each factor rated on a scale from
0 to 10 with ten being good

 Clarkston Richland Portland

Employee satisfaction factors

Real-estate 0.00082 0.006816 0.000468
availability
(housing permit
ratio to
population
higher the
better)

Real-estate $66,100 $221,600 $169,700
cost (single
family new
housing
construction
permit avg.
cost '04)

K-12 Schools High High Well over
 Schools: Schools: 10 in each
 1 public 3 public, public and
 1 private private
 in all
 Primary/ Primary/ three
 Middle Middle areas of
 Schools: 8 Schools: education
 public, 1 10 public,
 private 2 private

Culture Limited Diverse Highly
Availability Diverse

Both Employee Satisfaction and Plant Functionality
Factors

Local & State Tax sales tax but sales tax but No sales tax,
Rate low income low income but high
 tax tax income tax

Airport Flights Flights Portland
Availability available to available to International
 international international Airport
 airports airports

University University of University of University
Education Idaho/ Idaho/ of
Availability Washington Washington Portland/
 State State Portland
 University/ University/ State
 Lewis and Columbia University/
 Clark State Basin College Concordia
 College University

Plant Functionality Factors

Relevant to University of University University
biodiesel research Idaho/ of Idaho/ of
University Washington Washington Idaho/
 State State Washington
 University University State
 University/
 Oregon State
 University

Labor availability 6.30% 5.60% 6.20%
(unemployment rate,
assume same pay rate
at all sites)

Established Network High Low None
with Farmers
(80 mile radius)

 Total

Factors not included as they did not vary or were
not important

Labor skill level 81.40% 92.60% 85.70%
(35 years old
% graduated high
school)

Hospital Kadlec Tri-State Approx. 8
 Medical Memorial including
 Center and Hospital and OHSU
 Lourdes St. Joseph hospitals and
 Counseling Medical clinics and
 Center Center Doernbecer
 (5 miles)

Hospital treat chem. Yes Yes Yes
Exposure

Site Availability Adequate Adequate Adequate
 growth growth growth
 potential potential potential

Diversity 93% White 87% White 75.5% White
 non-Hispanic non-Hispanic non-Hispanic

Production Growth Unlimited Unlimited Unlimited
Possibilities

Feedstock Growth Unlimited Unlimited Unlimited
Possibilities

Local Demand growth Stable Stable Some
 Growth

 Clarksto Richlan St.

Employee satisfaction factors

Real-estate 8 10 5
availability
(housing permit
ratio to
population
higher the
better)

Real-estate 10 5 7
cost (single
family new
housing
construction
permit avg.
cost '04)

K-12 Schools 7 9 10

 9 9 10

Culture 7 9 10
Availability

Both Employee Satisfaction and Plant Functionality
Factors

Local & State Tax 9 9 7
Rate

Airport 9 9 10
Availability

University 10 8 10
Education
Availability

Plant Functionality Factors

Relevant to 10 8 5
biodiesel research
University

Labor availability 10 8 10
(unemployment rate,
assume same pay rate
at all sites)

Established Network 10 0 0
with Farmers
(80 mile radius)
 99 84 84

Factors not included as they did not vary or were
not important

Labor skill level 8 10 9
(35 years old
% graduated high
school)

Hospital 10 10 10

Hospital treat chem. 10 10 10
Exposure

Site Availability 10 10 10

Diversity 7 8 9

Production Growth 10 10 10
Possibilities

Feedstock Growth 10 10 10
Possibilities

Local Demand growth 9 9 10

 Weighted Rating

 Weight Clarksto Richlan St.

Employee satisfaction factors

Real-estate 0.05 0.4 0.5 0.3
availability
(housing permit
ratio to
population
higher the
better)

Real-estate 0.05 0.5 0.3 0.4
cost (single
family new
housing
construction
permit avg.
cost '04)

K-12 Schools 0.03 0.2 0.3 0.3

 0.03 0.3 0.3 0.3

Culture 0.05 0.4 0.5 0.5
Availability

Both Employee Satisfaction and Plant Functionality
Factors

Local & State Tax 0.10 0.9 0.9 0.7
Rate

Airport 0.07 0.6 0.6 0.7
Availability

University 0.07 0.7 0.6 0.7
Education
Availability

Plant Functionality Factors

Relevant to 0.10 1 0.8 0.5
biodiesel research
University

Labor availability 0.15 1.5 1.2 1.5
(unemployment rate,
assume same pay rate
at all sites)

Established Network 0.30 3 0 0
with Farmers
(80 mile radius)
 1 9.5 5.8 5.8

Factors not included as they did not vary or were
not important

Labor skill level 0.00 0 0 0
(35 years old
% graduated high
school)

Hospital 0.00 0 0 0

Hospital treat chem. 0.00 0 0 0
Exposure

Site Availability 0.00 0 0 0

Diversity 0.00 0 0 0

Production Growth 0.00 0 0 0
Possibilities

Feedstock Growth 0.00 0 0 0
Possibilities

Local Demand growth 0.00 0 0 0