Optimizing Transportation Costs For Goodman Fielder NZ

Background

Transportation costs form a significant part of the total operational costs of a manufacturer. Finding an optimum solution to the case company is a food manufacturing company which specializes on loafs, groceries, daily products and oils among others. The company operates four bakeries that supply loaf to the Northern Island of New Zealand. The goal of the current analysis is to demonstrate how linear programming can be used to minimize the transportation costs for bread supplied to the Island. To do this, the analysis shall seek to answer the following questions:

RQ1: Does the company have the baking capacity to meet the annual demand for its breads in New Zealand’s Northern Island?

RQ2: If not, how should the company increase its production capacity to meet the demand?

RQ3: If yes, what amount of bread should be transported from each bakery to each demand center to minimize the transportation costs?

In today’s increasingly globalized and competitive business environment, cost management is an important determiner of corporate profitability and sustainability. Cost management involves a variety of approaches and strategies designed to lower the total costs incurred in the production of goods and services (Bhimani, 2018). To achieve this, measures must be taken to reduce the costs incurred at each stage in the supply chain. Some of the common supply chain management approaches to minimizing costs include lean manufacturing, scheduling, and inventory management (Oluwagbemiga, Olugbenga, & Zaccheaus, 2014). This report explores one of the most robust and commonly used approaches to minimizing distribution costs, transportation costs and operational costs in general.

Transportation account for a significant cost of doing business for most contemporary businesses. This is particularly important for companies that source bulk inputs from or supply bulky goods to distant locations. For such businesses, there is a significant potential for cost cuts through optimization of the transport network (Jackson, 2015). This case study explores the potential benefits of applying linear programming techniques to minimize bread transportation costs for Goodman Fielder NZ.

Goodman Fielder NZ is a subsidiary of the Singapore and Hong Kong owned food processor and wholesaler Goodman Fielder. The company specializes in pastries including loafs and groceries. In New Zealand, the company operates 13 bakeries which supply the country. Six of these bakeries are located in the Northern Island. The six bakeries are located at different parts of the Island and supply the entire Island (Goodman Fielder NZ, 2018). The demand of bread is typically highest in urban centers; in this case 12 cities in the Island (Ibisworld, 2017). The current analysis sought to explore how the company can optimize its transportation network to reduce costs. This will improve the company’s profitability, sustainability and the company’s competitive advantage in the increasingly competitive market.

Linear Modeling for Transportation Problem

The New Zealand baking industry recorded $503 million in sales for the 2017 financial year. In the last five years, the industry has recorded a decline in the market size at an average -1.5%. The industry comprises of more than 1240 business. A majority of these business are small scale bakeries including local hot bread shops and family run bakeries (Deloite, 2018). Such businesses mostly bake and sell their products to end consumers at the same premises. However, there is a glowing trend where customers increasingly prefer pre-packaged bread (pastries). Typically, pre-packaged breads are manufactured by the few large plant bakers who distribute them to supermarkets and other types of shops who then retail them to the end consumer (Australia and New Zealand Banking Group, 2001). Goodman Fielder is a large scale baker.

Goodman Fielder is a Singapore and Hong Kong owned food processer and wholesaler. The company specializes in the production of based products including loaves, cakes and other pastries, dairy products, oils and other foodstuffs. Unlike smaller bakeries who mostly sell all of their products themselves, Goodman Fielder is a wholesaler. The company sells its products to retailers who then sell to the ultimate consumers (NZ Ministry of Business, Innovation and Employment, 2013).

The company’s business model has a few advantages and disadvantages. First, due to its large scale capacity, the company enjoys the benefits of economies of scale. However, given its supply chain, the final price of the bread includes the retailer’s premium (Duan, 2010). As such, if all production costs are constant, the company’s loaves would sell at higher price than those produced by small scale producers who do not sell via a middle man. The company can compensate on this by exploiting its capacities to minimize other production costs. This will allow competitive pricing and, hence, enhance profitability and competitive advantage.

As mentioned previously, the company operates six production plants in the Northern Island (Goodman Fielder NZ, 2018). Figure 1 below show the locations of the plants. The Puhoi Valley Cheese and Earnst Adams Palmerston North plants specialize in other products other than bread. Consequently, only the other four production plants are considered as bread bakeries in this analysis.

 

Figure 1: Goodman Fielder Production Plants

The four bakeries supply the entire Island with the company’s bread with demand being concentrated at the major urban centers. For simplicity, demand is considered to originate only in the 12 major Northern Island cities. Figure 2 below shows the locations of these major demand locations. Typically, the company transports the bulk loaves of bread to multiple retailers in the demand centers. In this analysis, the demand is treated as originating from all retailers in a certain city and its environs is treated as originating from a single point in the respective city (Kant, 2014). Typically, the company uses its specialized trucks to deliver the products; and outsource transportation whenever demand exceeds capacity.

Data and Assumptions

 

Figure 2: Main Bread Demand Locations in New Zealand

Linear modeling is used to find a solution to Goodman Fielder’s transportation problem. The goal is to optimize the transportation network and resources to minimize costs while meeting all of the demand. The company does not provide elaborate data on it operations that is required to conduct the analysis. Consequently, estimations were used to generate the data required.

First demand forecasts are an important input into the transportation planning. First, it is assumed that retailers order the company’s products in truckloads. There is a strict requirement that orders should never be a fraction of a truckload. Estimates of the demand per demand center are shown in table1 below.

Demand Center	Demand (Truckloads)
Paihia	200
Whangarei	60
Auckland	400
Hamilton	150
Tauranga	160
Rotorua	100
Gisborne	70
Taupo	200
Napier	120
New Polymouth	190
Palmerstone North	280
Wellington	700
Total	2630

Table 1: Fielder Bread Demand

The ability to meet the demand is directly influenced by the capacities of the bakeries. In this case, there are two limiting factors. First, the amount of bread that each bakery can produce is limited to a given value. Next, each bakery has a transportation department with a given capacity to transport the products. This transportation constraint is catered for by limiting the amount of distance that each department can transport the loaves of bread annually. The annual production capacity and the distance capacity estimates used in this analysis are presented in table 4 below.

Bakery	Truckload	Distance Capacity (KM)
Auckland	600	88000
Huntly	600	88000
Hawke’s	600	95000
Wellington	1100	130000
Total	2900	401000

Table 2: Bread Production Capacity

Next, estimates of the costs and distances involved in transporting the bread from the bakeries to the various demand centers are required to solve the transportation problem. Table five below summarizes the estimated costs for transporting a truckload of loaves for one kilometer and the total distance between the bakeries and the demand centers. Notably, some routes (path from bakery to demand center) are very long such that transportation along these routes is not cost effective under any scenario. Therefore, table five below includes estimates for the routes whose costs are tolerable.

Route (Bakery-Demand Center)	Cost Per KM ($)	Distance (KM)
Auckland-Paihia	520	100
Auckland-Whangarei	1020	200
Auckland-Auckland	270	50
Auckland-Hamilton	1020	200
Auckland-Tauranga	1520	300
Auckland-Rotorua	1770	350
Huntly-Paihia	670	130
Huntly Whangarei	1120	220
Huntly-Auckland	420	80
Huntly-Hamilton	770	150
Huntly-Tauranga	920	180
Huntly-Rotorua	1020	200
Huntly-Gisborne	1520	300
Huntly-Taupo	1270	250
Huntly-Napier	1770	350
Huntly-New Polymouth	1020	200
Hawke’s-Tauranga	1670	330
Hawkes-Rotorua	1370	270
Hawke’s-Gisborne	1070	210
Hawke’s-Taupo	870	170
Hawke’s-Napier	770	150
Hawke’s-New Polymouth	1020	200
Wellington-Rotorua	2020	400
Wellington-Gisborne	1570	310
Wellington-Taupo	1620	320
Wellington-Napier	1520	300
Wellington-Palmerston North	520	100
Wellington-Wellington	270	50

Table 3: Route Information

The objective of the analysis is to minimize the total bread distribution costs incurred by Goodman Fielder given the situation described by the data in the previous section. Here, the transportation costs are a function of the distance which the goods are transported and the number of truck loads transported along each route. The objective function is given by:

TC=TL1*C1+TL2*C2…….

Where TC=Total distribution costs

TLi=the number of truck loads transported along route i

Ci= transportation cost per truck load alongside route i

Objective Function and Constraints

The goal of the analysis is to find the values of TLi for all of the routes that will ensure that all demand is met at the lowest distribution costs possible (minimize TC) (Ghazali, Majid, & Shazwani, 2012).From the data section, the minimization problem is subject to the following constraints.

a) Demand

Failure to deliver all of the loaves demanded is undesirable. This translates into loss of potential earnings, customer trust and market share (Patel, 2017). Therefore, the breads supplied to each demand center must be equal to the demanded amount.

A bakery cannot produce more bread than it has the capacity to do. Therefore, all bread supplied from a given bakery must be equal or less than its production capacity.

It is assumed that each bakery has a transportation department with a fixed capability. Each transportation department can only transport loaves produced at the bakery where it is based. Therefore, the total distance traveled in distributing bread from a given bakery is constrained by the bakery’s transportation department capacity.

The transportation costs have a fixed component as well as a variable component which is directly related to distance. Consequently, the transport cost per unit is lowest when trucks carry the maximum capacity. Consequently, it is assumed that no retailers order portions of truck loads; or that any such orders are combined with others to make a truckload. Therefore, each truck that leaves a given bakery carries a full truckload. In the model, this constraint is achieved by a constraint requiring that all TL values should be integers.

TL=Int

Finally, a constraint is put in place to ensure that the solver algorithm does not set negative numbers of truckloads of bread to be delivered. A negative TL is impractical in this case since it would imply that bread is moving from the retailer to the wholesaler. Consequently, each TL value is constrained to positive values using the following constraint:

TL-Absolute (TL) =0

Capacity and Demand

The first research question sought to establish whether the four bakeries have the capacity to meet the current demand. In total, the demand centers require 2630 truckloads of bread per month. The four bakeries have a capacity of producing 2900 truckloads of bread. Consequently, the current production capacity is sufficient to meet the total demand.

Table five below summarizes the optimum solution obtained for the transportation problem. Specifically, the table shows the number of truck loads that should be transported via each route as well as the total costs incurred for using the route and the total distance covered on the route.

	Truckloads	Cost Per KM/Truckload	Distance Per Trip(KM)	Total Costs Per Route	Distance Per Route
Auckland-Paihia	200	520	100	103999	20000
Auckland-Whangarei	0	1020	200	2	0
Auckland-Auckland	400	270	50	107999	20000
Auckland-Hamilton	0	1020	200	0	0
Auckland-Tauranga	0	1520	300	0	0
Auckland-Rotorua	0	1770	350	5	1
Huntly-Paihia	0	670	130	2	0
Huntly Whangarei	60	1120	220	67197	13200
Huntly-Auckland	0	420	80	1	0
Huntly-Hamilton	150	770	150	115500	22500
Huntly-Tauranga	160	920	180	147200	28800
Huntly-Rotorua	100	1020	200	101997	19999
Huntly-Gisborne	0	1520	300	0	0
Huntly-Taupo	0	1270	250	0	0
Huntly-Napier	0	1770	350	39	8
Huntly-New Polymouth	17	1020	200	17813	3493
Hawke’s-Tauranga	0	1670	330	0	0
Hawkes-Rotorua	0	1370	270	0	0
Hawke’s-Gisborne	43	1070	210	46480	9122
Hawke’s-Taupo	196	870	170	170796	33374
Hawke’s-Napier	120	770	150	92383	17997
Hawke’s-New Polymouth	173	1020	200	175987	34507
Wellington-Rotorua	0	2020	400	0	0
Wellington-Gisborne	27	1570	310	41700	8234
Wellington-Taupo	4	1620	320	5966	1178
Wellington-Napier	0	1520	300	1	0
Wellington-Palmerston North	280	520	100	145600	28000
Wellington-Wellington	700	270	50	189000	35000

Results and Analysis

Table 4: Optimum Solution

Table six below summarizes the demanded volumes of bread and the volumes that will be supplied under the optimal solution in truckloads. From the table, each of the demand centers will receive the exact amount of read that is demanded. This shows that the demand constraint is met and not demand will be left unsatisfied.

Demand Center	Demanded (Truckloads)	Supplied (Truckloads)
Paihia	200	200
Whangarei	60	60
Auckland	400	400
Hamilton	150	150
Tauranga	160	160
Rotorua	100	100
Gisborne	70	70
Taupo	200	200
Napier	120	120
New Polymouth	190	190
Palmerstone North	280	280
Wellington	700	700
Total	2630	2630

Table 5: Demand Constraint

Finally, table 7 below summarizes the data on the capacity (production and transportation) for each of the bakeries. From the table, it is observed that the optimal solution uses lesser resources than the capacities of the production plants. This implies that no bakery will be overstretched or require any expansion.

Bakery	Truckloads Capacity	Truckloads Delivered	Distance Capacity (KM)	Distance Covered (KM)	Capacity Utilization
Auckland	600	600	88000	40001.10	1.00
Huntly	600	487	88000	88000.00	0.81
Hawke’s	600	532	95000	95000.00	0.89
Wellington	1100	1010	130000	72412.40	0.92
Total	2900	2630	401000	295413.50	0.91

Table 6: Capacity Constraint

The current analysis has one major limitation. This is the fact that fictitious data is used to test the linear model. The case company does not provide detailed data regarding its demand estimates, capacity and transportation costs. Therefore, the current solution is as accurate as the data estimates are. While the solution provides a suitable illustration of the application of linear modeling to transportation problems, it is not an accurate representation of the situation for Goodman Fielder.

From the foregoing, linear programming can be applied to minimize transportation costs while meeting demand and capacity requirements. In light of this, Goodman Fielder (and companies in the industry in general) should adopt linear programming approach to optimize their transportation. The costs cuts realized should be passed down to consumers through competitive prices (Sulanjaku, 2015). This will ultimately enhance a company’s competitive advantage (Wagner, 2008).

Auckland Regional Coincil. (2010). Industry snapshot for the Auckland region: Food and Beverage Sector. Auckland : Auckland Regional Coincil.

Australia and New Zealand Banking Group. (2001). INDUSTRY BRIEF: ANZ Food and Beverage Industry. Sydney: Australia and New Zealand Banking Group.

Bhimani, A. (2018). Cost management in the digital age. LSE Research Online.

Deloite. (2018). New Zealand’s food story: The Pukekohe Hub. Wellington: Deloite.

Duan, Y. (2010). Buyer–vendor inventory coordination with quantity discount incentive for fixed lifetime product. International Journal of Production Economics, 351-357.

Ghazali, Z., Majid, A. A., & Shazwani, M. (2012). Optimal Solution of Transportation Problem Using Linear Programming: A Case of a Malaysian Trading Company. Journal of Applied Sciences , 2430-2435.

Goodman Fielder NZ. (2012). The 2012 Goodman Fielder Sustainability Report. Wellington: Goodman Fielder NZ.

Goodman Fielder NZ. (2018). Locations. Retrieved from Goodman Fielder NZ: https://goodmanfielder.com/careers/locations/

Ibisworld. (2017). Bakery Product Manaufucturing-New Zealand Market Research. IBISWorld.

Jackson, J. (2015). QUANTITY DISCOUNTS, CAPACITY DECISIONS, AND CHANNEL CHOICES . New York: New York State University.

Kant, G. (2014). Coca-Cola Enterprises Optimizes Vehicle Routes for Ef?cient Product Delivery. Interfaces.

Kostoglou, V. (2018). Quantity Discount Model Cas Studies and Solutions. Retrieved from https://aetos.it.teithe.gr/~vkostogl/files/Educational%20material/Linear%20Programming_case%20studies+solutions.pdf

NZ Ministry of Business, Innovation and Employment. (2013). An Investor’s Guide to New Zealand Food & Beverage Industry 2013. Wellington: Ministry of Business, Innovation and Employment.

Oluwagbemiga, O. E., Olugbenga, O. M., & Zaccheaus, S. A. (2014). Cost Management Problem and Firm’s performance of Manufucturing Companies. International Journal of Economics and Finance.

Patel, R. G. (2017). Optimal Solution of a Transportation Problem. Global Journal of Pure and Applied Mathematics.

Sulanjaku, M. (2015). STRATEGIC COST MANAGEMENT ACCOUNTING INSTRUMENTS AND THEIR USAGE IN ALBANIAN COMPANIES. European Journal of , Economics and Accountancy.

USDA Foreign Agricultural Service. (2018). New Zealand Food Processing Ingredients Report 2018. Wellington: USDA Foreign Agricultural Service.

Wagner, S. (2008). Cost management practices for supply chain management: an exploratory analysis. International Journal Services and Operations Management.