See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/242548206

Using VAT Analysis as a Framework for Supply Chain Management: A Case Study

Article in European Journal of Economics, Finance and Administrative Sciences · March 2009

CITATIONS READS
0 1,253

2 authors, including:

Archie Lockamy
Samford University
45 PUBLICATIONS 2,336 CITATIONS

SEE PROFILE

All content following this page was uploaded by Archie Lockamy on 12 October 2017.

The user has requested enhancement of the downloaded file.

European Journal of Economics, Finance and Administrative Sciences
ISSN 1450-2887 Issue 17 (2009)
© EuroJournals, Inc. 2009
http://www.eurojournals.com

Using V-A-T Analysis as a Framework for Supply Chain

Management: A Case Study

Michael S. Spencer
College of Business Administration, University of Northern Iowa
E-mail: [email protected]
Tel: 319-273-6793; Fax: 319-273-2922

Archie Lockamy
Samford University School of Business
E-mail: [email protected]
Tel: 205-726-4135

Abstract
Just-in-Time methods are used by managers taking a systems-wide view of production
planning and control that stretches beyond organizational boundaries. The systems-wide
view is often called Supply Chain Management when a producer looks beyond the
manufacturing organization into the supplier base and also towards end customers. This
view can extend through the entire logistics channel. Many researchers have commented on
the need to manage the entire supply chain under JIT rather than focusing on a single
channel member organization. However, there appears to be no coherent framework from
which to examine a supply chain. Theory of Constraints provides a framework called V-A-
T analysis that could be used to provide a channel-wide framework. This article examines
the use of V-A-T analysis in a supply chain via a case study, and proposes a research
agenda for using TOC methods.

Keywords: Supply chain management, theory of constraints, V.A.T. structures

1. Introduction
The purpose of this paper is to first illustrate the applicability of the Theory of Constraints' V-A-T
framework to a supply chain and then propose a research agenda using Theory of Constraints (TOC)
methods for improving overall supply chain system performance. First, we will discuss why a systems-
wide perspective has developed, and thus the need for a research framework. Next, the V-A-T
framework will be presented. A case study will then be used to illustrate the applicability of V-A-T as
a framework for understanding a supply chain. Finally, a series of propositions concerning TOC's
application to supply chain management will be raised for further research.
One problem in conducting research on the supply chain by production management
investigators is that there appears to be no coherent SCM framework from which to develop a research
program. A review of the literature found that most supply chain models were simply linear
representations. There did surface a few good conceptual diagrams (Handfield & Nichols, 1999;
Monczka, Trent & Hanfield, 1998; Dobler & Burt, 1996) of a supply chain framework but no overall
model from which hypotheses could be developed from field studies. The logistics field also lacked an
70 European Journal of Economics, Finance And Administrative Sciences - Issue 17 (2009)
overall model that integrated production management concepts although more generalized models were
found (Bowersox, et al., 1986; Stock & Lambert, 1997). One research text (Bowersox et al.; 1992) did
propose a useful model and a series of hypotheses concerning the systems view of logistics but lacked
any production management details. To illustrate the true complexity of a supply chain, one text
illustration of a supply chain depicted 19 automotive assembly plants servicing a large dealer network,
being supplied by 57 manufacturing plants that, in turn, were being supplied by hundreds of second tier
suppliers (Handfield & Nichols, 1999). The commonly used linear models, e.g. (raw material ->
manufacturer -> retailer) simply were unable to depict the complexity of a real supply chain.
TOC researchers have developed a framework which describes a complex production system in
general. This framework also has the added benefit of tracing its origins to the MRP system components
of routings and bills of materials. The framework is called the V-A-T logical structure. The name
originates from its MRP roots where Goldratt (1990) combined the product bill of material with the
component part's operation routings. The term "V-A-T" comes from the shape of the resulting three
diagrams. V-A-T is used to position the buffers and control points used in the TOC scheduling method
called Drum-Buffer-Rope and apply other TOC methods to the production process. The V-A-T logical
structure framework has now been used in a number of varied applications from focusing JIT methods
(Lockamy & Cox, 1991) to describing warehouse configurations (Spencer, 1993) and optimizing new
product introductions (Nels & Plenert, 1993).
However, it is unclear from the TOC literature how to apply V-A-T to a complex system
consisting of separate organizational units. Goldratt (1990) appears to propose that there will be only one
constraint for a system no matter how complex. Others propose that there are multiple V-A-T structures in
a complex system, each having its own constraint (Umble and Srikanth 1990). Further, the application of
TOC to SCM is also unclear. Some proposing that a one-for-one (lot-for-lot) replenishment strategy
throughout the entire supply chain be used to avoid delivery fluctuations and stockouts (Blackstone &
Cox, 1998).

2. A System-wide Perspective of Production

The integration of Just-in-Time (JIT) methods into production management may have had a profound
organizational impact. Once thought of as a method to reduce inventory and, later, as a method to
better manage the flow of inventory, the more subtle managerial aspects of JIT are being researched.
JIT's success forces management to take a more systems-wide view of the production process rather
than to see production or materials management as an independent function operating within the
business.
The origin of this change may be traced back to the advent of Manufacturing Resource
Planning, (MRPII in production literature). Practitioners seeking to implement the closed-loop MRPII
approach found themselves crossing organizational boundaries in order to accomplish business
objectives. Feedback from purchasing to master production scheduling (MPS), and feedback from
accounting to production planning, was coupled with feedback from MPS to marketing to provide a
more integrated approach to production management. This systems-wide approach was not as well
received by the other business functions, nor even by materials professionals, as might have been
expected. Materials professionals found that there were reasons why organizational boundaries existed
other than for purely profit-maximizing purposes.
What seemed to be drawing materials managers into the cross-functional view was made more
visible by the emergence of TOC popularized in the book The Goal, by Goldratt (1984). A production
system was best seen as a system of interdependent operations with a single constraint limiting overall
throughput. Managing the constraint ensured maximizing the system's throughput. Production literature
began to discuss first the superiority, then the relationship among the three production planning
methods: MRPII, JIT, and TOC. Still missing from the discussion was the application of these methods
outside of traditional manufacturing. It should be noted that Goldratt did raise the issue in later works
71 European Journal of Economics, Finance And Administrative Sciences - Issue 17 (2009)
by stating that a constraint could lie outside the traditional domain of production management as a
logistical constraint or as a managerial policy constraint (Umble & Srikanth, 1990).
An article in the 1992 APICS International Conference proceedings (Burnham, 1992) appears
to be one of the first production researchers to articulate the need for broadening the materials
management perspective to view the overall logistics system in order to gain the full benefits of JIT.
Burnham, also an earlier researcher in JIT practices, found key JIT concepts could be applied to the
entire logistics system. He identified the goal of the logistics management as "...to bring materials from
the lowest-tier suppliers through the various conversion processes and into distribution to reach the
customer (p.500)". Burnham claimed that the material flow through the logistics channel could be
managed and improved using the same JIT methods as applying JIT to a single factory. Burnham
called this the Global Logistics System (GLS) addressing both strategic as well as tactical issues and
tradeoffs. Other researchers also identified the need to take a systems-wide perspective (Toni, et al.,
1992; Novack, et al., 1993; and Scully & Fawcett, 1993) as well.
Proceedings of recent APICS International Conferences do reflect an on-going discussion
among practitioners and consultants concerning integrating the traditional materials management topics
into a systems-wide framework, e.g., (Lehmann, 1996; Rosen, 1996; Minsky, 1996; Mlot et al., 1996;
and McMullen, 1996; Alber & Walker, 1997, and Walker, 1999) with the term "supply chain
management" (SCM) gaining acceptance rather than GLS.
Theoretically, TOC is applicable to any system (AGI, 1995, p.3). Thus, there should emerge a
constraint within the supply chain system which can then be better managed using TOC methods with
an overall improvement in the SCM system performance. This paper will focus on the use of the VAT
framework for examining a supply chain system, thus facilitating the application of TOC tools such as
drum-buffer-rope to improve supply chain performance similar to results reported in the TOC literature
on their application to a single entity within a supply chain.

3. Methodology
This research is exploratory in nature and will follow the methods suggested by Marshall and
Rossman, (1989) Patton (1990), and Strauss (1990) for qualitative case-based research. This is part of a
larger research effort where a logistics channel was selected to be examined. The channel consisted of
a JIT manufacturer, a smaller supplier company, the supplier's raw material providers, and the
interconnecting transportation system. This paper will examine only the supplier company and its
relationship to the customer and raw materials supplier. The relationship will be examined, as
suggested by Handfield & Nichols (1999) by developing the information flows that correspond with
the materials flows from customer order to finished component delivery and then testing the
applicability of using the VAT framework rather than linear models.
A series of interviews were conducted to gather information concerning the production and
purchasing planning and control functions at John Deere Waterloo Operations and at the supplier
company, the GMT Corporation. The next section presents the companies' backgrounds, processes, and
information flows.

4. The Deere-GMT Supply Chain

The John Deere Company is the world's leading manufacturer of heavy-duty farm equipment such as
two-wheel row crop tractors, larger four-wheel drive tractors and combines. The company's primary
heavy equipment manufacturing facilities are in Waterloo, Iowa (for tractors) and Moline, Illinois (for
combines) in the United States, and in Mannhiem, Germany and in several locations in France. Overall
the company had sales of approximately $10B (US) in 1999. Tractor sales account for about 25% of
the total.
72 European Journal of Economics, Finance And Administrative Sciences - Issue 17 (2009)
Prior to industry consolidation in the 1980s, Deere prided itself in producing the vast majority
of its components. Since the company has been a major producer of farm equipment since the 1920s,
the production facilities developed over a long period of time and are located in multiple-story
buildings spread over a large area. Factories were dedicated to specific types of vehicles and as such
fabrication and assembly was conducted within a single complex. The company's largest single
complex is the Waterloo Works which once produced all of the company's row crop and four-wheel
drive tractors. As a result of JIT implementation and cost pressures the company now out-sources a
substantial portion of the fabrication operations to supplier companies. The down-sizing process has
reduced the overall manufacturing floor space used at the original Waterloo Works site by more than
60%.
In addition, a state-of-the-art tractor assembly factory went into production in Waterloo, IA in
1982 located about 5 miles from the original manufacturing complex as the major user of components.

4.1. The Waterloo Factory

The John Deere Waterloo Works was the original site of farm tractor production beginning in the
1920s. The word "works" is used by Deere to describe a production complex consisting of many
separate factories or facilities. At one time, the Waterloo Works consisted of two separate foundries,
two large single story production buildings, six major multiple-story manufacturing buildings and over
20 smaller production buildings. The John Deere Waterloo Works once employed over 17,000 salaried
and wage employees and produced over 165 tractors per day. With the relocation of diesel engine
production and tractor assembly to new sites, and the out-sourcing program, current employment is less
than 5000, supporting production of approximately 107 tractors per day.
The Waterloo Works is located next to the John Deere Electric Foundry. The Waterloo Works
conducts itself as a separate business with all components sourced by a bidding process. A separate
labor contract with the United Auto Workers exists with a lower wage scale than the remaining
Waterloo Works complex. Even though the Electric Foundry is located adjacent to the Waterloo
Works, there is no guarantee that a component will be sourced there. In fact, considerable out-sourcing
of foundry castings has occurred as a result of the bidding process.
Likewise, the John Deere Tractor Assembly factory utilizes the same bidding process to source
its components. There is no guarantee that a component currently produced at the Waterloo Works will
continue to be purchased by the Tractor Division, unless successful in the bidding process. Component
price, quality and delivery reliability are three major criteria in the bidding process.

4.2. The Transmission Department

The transmission department manufactures cast iron cases called "casings" for the tractor
transmissions. The transmissions are then assembled using the casing and other manufactured and
purchased components in another area within the department. The transmission casings measure
approximately three feet wide, three feet high and vary from two feet to four feet long. Casings weigh
between 100 to 150 pounds. The Transmission department manufactures and assembles 23 different
casings at about 135 transmissions per day for both production and service.
The average routing for a transmission case is six machine centers, although many operations
are performed at a single machine center. Average set-ups between major part families range from two
to four hours. There are currently 24 wage employees assigned to the department. The castings for the
transmission cases are sourced from either the John Deere Electric Foundry, or from an outside
independent foundry.
The next section focuses on the production planning and control methods used at the
transmission manufacturing and assembly department at the Waterloo Works as an example of the
integration of JIT and MRP methods that allowed the development of a complex supply chain set the
stage for using VAT analysis.
73 European Journal of Economics, Finance And Administrative Sciences - Issue 17 (2009)
4.3. A Production Example
An example will illustrate the information and production flows for an internal Deere Waterloo Works
component. As orders for farm tractors are written for farmers by the John Deere dealer network, the
order, consisting of a due date and the various tractor options, is fed into the MPS at the Tractor
Assembly Division. The MPS is fed into the MRP system where net requirements for components are
calculated. For example, one component would be for a particular transmission. Each transmission has
a unique serial number that is tied to a specific tractor. The requirements are then sent into the inter-
factory system and transmitted to the supplying factory. In the case of a transmission, the requirement
is sent to the Waterloo Works. The Waterloo Works' MPS is then updated with the new requirement
with a corresponding due date to the tractor assembly build date. The MPS is then transmitted daily on
third shift to the Waterloo Works' MRP system.
The Waterloo Works MRP system explodes the transmission order through the bill-of-material
and calculates the net requirement for the transmission casing manufactured within the transmission
production area, for an assembly order in the transmission assembly area, and for all other components.
Additionally, the transmission casing requires a casting to be procured. The Waterloo Works'
MRP system then transmits the net requirement for the transmission casing casting to the purchasing
system to either the Deere foundry or an outside foundry depending on previously determined
percentages. In the event that a component will requires operations at an outside supplier, there will be
a parallel purchase order triggered in the purchasing system. This situation will be discussed in the
following section on GMT.
Since the transmission casings are procured from both the Electric Foundry and an outside
vendor, the purchasing system selects, based on predetermined constraints, the amount of casting
required from the outside vender and the amount needed from the Electric Foundry as well as the
corresponding due dates. Since the Electric Foundry is a Deere factory, those requirements are
automatically entered into the inter-factory system and transmitted into the Electric Foundry's MPS.
The MPS feeds its requirements into the foundry's MRP where production schedules are altered as
required to meet the calculated production due dates.
Execution of the production plan is problematic. Changes can occur throughout the production
planning cycle. For example, customers may change the configuration of an ordered tractor, in addition
to the normal changes seen in an MRP system. The execution of the production requirements requires
manual intervention.
For example, within the transmission department the changeover times vary from two to four
hours depending on the family of parts being manufactured. Within a family of parts the changeovers
are relatively small. However, going from one part family to another requires a major changeover
tending to the longer times. In addition, there are requirements for preventive maintenance, as well as
experimental production and unplanned machine breakdowns that occur. A production scheduler is
assigned to manually review all the requirements from the MRP system and develop supporting
production schedules. It is these manual schedules that are executed.
A similar series of activities are required at the foundry. The economics of foundry operations
require that considerable batching of casting requirements occur. Foundry pours are unique and
somewhat unpredictable. Sometimes the pour of iron into molds goes smoothly while at other times
there is considerable difficulty. Additionally, the chemistry of each pour must be determined as early
as possible in order to maintain quality throughout the pour. As such, a production scheduler manually
determines the foundry pour schedule using the net requirements from the MRP system.
At the Waterloo Works the transmissions are assembled along an assembly line which is similar
in operation to the tractor line described above. Kanbans trigger replenishment from the staging areas
at the Waterloo Works supplied from the machining area.
Components for transmissions that are manufactured by outside vendors, such as GMT, flow
into a warehouse and are then sent to the transmission assembly area as needed.
74 European Journal of Economics, Finance And Administrative Sciences - Issue 17 (2009)
4.4. The Example with an Outside Supplier-GMT Company
GMT is a privately-owned company operating in the Midwest. The company began as a tool and die
shop in 1973. GMT gained a reputation for high quality at a good price in the tool and die business for
several large manufacturers in the surrounding states. As the economy shifted and JIT was introduced
in the larger manufacturers, opportunities arose to add machining component production to the tool and
die business. As more production was outsourced, the majority from the John Deere Waterloo Works,
this segment of GMT grew to dominate the original company business.
The internal performance measurement system used at GMT for Deere components consists of
the following:
Quality - GMT is currently certified by Deere, Case, Sundstrand etc. They are currently
investigating ISO 9000 certification.
Inventory -Work-in-Process turnover is currently 8 to 10 times per year. WIP accuracy is a
concern. GMT uses a cycle counting inventory accuracy method with an annual
physical inventory. Currently there is about $165,000 of non-active Deere components
inventory on hand.
Shipping - target is 2 days early to Deere and 0 days late, they are achieving 98% on-time delivery
performance.
MIS - The management information system for purchasing and production planning consists
of two separate computer systems: The "Globe" system for Maintenance and Repair
purchases and "Profitkey" an off-the-shelf MRP system used for internal financial
controls and purchasing and production installed 1988.

4.5. The (CNC) Production Shop

The production shop, called the CNC (computer numerically controlled) shop at GMT, produces in
small to large size runs ranging from 20 to 35,000 pieces supplying Deere and others. The CNC shop
produces primarily turning work pieces with manufacturing capability up to 3 1/2 inch bar stock.
Manufacturing capability also consists of aluminum, steel, cast iron and stainless steel up to 1500 lbs
castings. There are about 100 direct labor employees assigned to 60 CNC machines and 39 machine
cells at GMT. The CNC shop has doubled in size over the past three years and is GMT's fastest
growing segment.
Both repetitive and batch production are occurring at the CNC shop with about 60% of volume
going to current production and 40% of production going to manufacturing service parts in batch. All
Deere components are shipped first to a local warehouse then either forwarded to factories, the Deere
Parts Distribution Warehouse, or directly to a dealership.

5. The GMT Production Information Flow

On a daily basis a GMT shop clerk reviews the Deere EDI mailbox and manually loads information
(for example, the delivery dates, quantities and part number for transmission components) into their
MRP system. Deere gives about a four month planning horizon to GMT via the EDI system. This
information originates from the Deere MRP system via the Deere Purchasing system. Most daily EDI
messages are changes to existing orders.
Each Friday, information is selected from GMT's MRP and entered into a stand-alone spread
sheet program. A GMT master schedule report is then generated by due date and by part number.
Production is then scheduled by a shop scheduler and shop supervisor on Saturday in a six hour
meeting. Each job is assigned to a work center using a spreadsheet. Work is scheduled by work center
for the next week. The MRP system is not used to develop the production schedules as GMT has not
yet decided on the usefulness of some of its MRP programs and reports.
75 European Journal of Economics, Finance And Administrative Sciences - Issue 17 (2009)
The CNC shop averages 150 to 200 jobs each week. Jobs are released to the shop using a
handwritten card that triggers material release. The job lot schedule is then loaded into MRP by the
release date. A hand written "Release for Work Orders" form is used to release material to the shop
along with traveler (route sheet) and blueprint. Raw material is delivered daily from the small GMT
warehouse based on this handwritten schedule.
When the job is complete the operator's "time ticket" is used to update a traveler and then the
MRP system is updated using the traveler by the shop clerk. The Deere EDI mailbox is then updated.
After production is run the MRP system is updated by the shop clerk loading the ship dates from
shippers.
Foundries have been advised at an earlier time period of the casting order by the Deere
purchasing system previously discussed.
Any casting shortages are given to the GMT buyer, by handwritten note "Material Shortage"
form. Expediting occurs by the buyer calling the supplier foundries and obtaining the production status
or by identifying the expected delivery date. Jobs are manually closed each week in the MRP system,
although this may require an inventory adjustment.
The GMT MPS goes to the material scheduler for materials planning. Each Friday the material
scheduler gets a list of all current WIP, a list of finished goods in inventory, and raw materials
quantities. From these he develops the materials plan using a spreadsheet. A spreadsheet is loaded with
the quantities and due dates, and is where an additional 4 to 5 weeks of lead time is added for raw
material by back scheduling from the due date.
From this spreadsheet a "Castings Order" report is prepared. The material scheduler makes an
educated guess about the correct WIP and makes adjustments to the spreadsheet based on current
production. These schedules are then updated daily and kept on file. The material scheduler tries to
schedule materials out 6 to 8 months based on minimum order size, foundry reliability, and unit price.
The GMT buyer receives a handwritten "Casting Order" sheet from the material scheduler each
Monday. This is reviewed and data manually entered into the MRP system. PO's (purchase orders) are
then generated and sent by fax to suppliers. These PO's are then matched at the suppliers with the
original PO's from the Deere purchasing system. Raw material is then sent from the foundry to GMT
using the Purchase Order due dates.
When raw material arrives at GMT a count sheet is prepared. The count sheet is sent to the
Buyer who updates the MRP system with inventory data. Production is then scheduled at GMT as was
previously discussed.
As a result of the implementation and use of MRP and JIT within the Waterloo Works, overall
inventory levels of transmissions, components, and casings decreased from 15 days on hand to 5.5 days
in the logistics channel excluding GMT's inventory. Transmission casings are valued at approximately
$400 to $600 (US) each, which complete transmission are valued at $4000 (US) each or higher.
Quality increased as measured by the defect rate and rework costs has declined approximately 25%
during the same time period.

6. Linear SCM Depiction

A review of the previous description of the information flows could now be used to depict the Deere-
GMT supply chain linearly as:
(Foundries-> GMT CNC Shop-> Deere Transmission-> Deere Tractor).
Applying TOC tools to the linear model, a constraint would be expected at, for example, GMT
which would require a buffer be established between the foundries and GMT. A communication device
(rope) would be established from the constraining work center at GMT causing the pace of production
throughout the supply chain to be set by the GMT work center. Material from the foundries would be
sent to GMT paced to the output of the GMT machine center constraint. A shipping schedule would be
76 European Journal of Economics, Finance And Administrative Sciences - Issue 17 (2009)
developed at the Deere Tractor Works based on the production received at the Deere Component
Works Transmission Department.
However, some problems now emerge with the linear approach. Not all parts for all
transmissions are manufactured at GMT. What is a non-GMT component machined elsewhere
becomes the constraint?
Not all tractors assembled at the Tractor Works use the same transmissions with the GMT
produced components. Some Tractors require transmissions made from components made by non-
GMT companies. Additionally, tractors have other major components such as engines and wheel
assemblies that, may themselves, have non-GMT components, but be subject to their own work center
constraint. Thus, the linear model lacks the ability to depict the richness and complexity of the true
supply chain. Application of TOC tools would then be difficult or, worse, be misapplied. A more
detailed depiction is required to identify the constraint(s) and their relationship to the entire supply
chain.
Based on the above information the chief characteristics of the supply chain consisting of
GMT, a major customer, John Deere Waterloo Works', GMT's foundry suppliers, and the Waterloo
Works' major customer (the Tractor Assembly factory) can be identified and compared with the
characteristics found in the V-A-T descriptions. First the proper logical structure must be identified
that is applicable to the entire supply chain. If that cannot be accomplished then the VAT logical
structure cannot be used lessening the likelihood of applying TOC tools to improve the system
performance.
The following are identified as general characteristics of the three logical structures (Umble &
Srikanth, 1990):

V-structures
1. The number of end items is large compared to raw material.
2. All end items sold are produced essentially the same way.
3. The equipment is generally capital intensive and highly specialized.
4. The key problem is the misallocation of material and over production.
5. Finished goods inventories are too high.
6. Customer service is poor.

A-structures
1. Large number of manufactured parts assembled into a small number of end items.
2. Component parts are unique to specific end items.
3. Routings for components are dissimilar.
4. Machines are general purpose.
5. The key problem is shortages in assembly.
6. Excessive overtime and wandering bottlenecks are observed.

T-structures
1. There are several common manufactured/purchased components assembled together to
produce the end item.
2. Component parts are common to different end items.
3. Component routings do not include divergent or assembly points.
4. Routings are dissimilar for components.
5. Key problem is the misallocation of components.
6. There are large finished good and component inventories.
77 European Journal of Economics, Finance And Administrative Sciences - Issue 17 (2009)
Figure 1: Typical product flow diagram for a v-plant, t-plant and an a-plant

a. V-Plant

assembly AR 1 AR 2 AR 3 AR 4

packing R3 R4

clipping R2

preparation R1

b. T-Plant

final assembly models X1 X2 X3 X4 X5

sub assembly AR6 AR7 AR8 AR

inspection opn. R1 R2 R3 R4

second opn. R1 R2 R3 R4

first opn. R1 R2 R3 R4

raw material R1 R2 R3 R4

c. A-Plant

final assembly X1

sub assembly AR 1 (6) AR 2 (1)

sub assembly R1 AR 4 R2 AR 3

second opn. R1 R3 R4 R2 R3 R5

first opn. R1 R3 R4 R2 R3 R5

raw material R1 R3 R4 R2 R3 R5

7. Discussion
According to Umble and Srikanth (1990) combinations of the structures also are possible such as a V
top and an A structure on the bottom. In a logistics channel this would mean that the each
organizational unit would have a separate constraint. Each of these constraints would have to be
managed using TOC methods. It is here that a problem with TOC methods seems to occur.
In the Deere-GMT supply chain we would expect to find a constraints at the Deere Tractor
Assembly factory described as a T-structure, another constraint at each of the supplying departments in
the Deere Waterloo Works, constraint at each of the suppliers to Deere such as GMT, and others at
each of the foundries. Since TOC focuses its attention on improving a system's constraint through the
78 European Journal of Economics, Finance And Administrative Sciences - Issue 17 (2009)
use of the TOC five step process (Goldratt & Fox, 1986), the logistics system would consist of many
interacting constraints.
An often heard criticism of JIT is that it forces inventory down the supply chain to smaller
suppliers. The use of this linear model seems to add legitimacy to this by suggesting that if TOC
methods are applied and each organizational unit is seen as a separate VAT structure, then the suppliers
would be forced to hold inventory as a constraint buffer positioned between the supplier and the large
manufacturer.
However, by using the VAT analysis to the entire supply chain a different framework emerges
in the case of Deere-GMT. The entire logistics system comprises a single A-structure.
By examining the above V-A-T characteristics and the information and production flows within
the Deere-GMT supply chain as an entire logistics channel the following results are obtained:
V-structure: Only characteristic 3, capital equipment, is true in some routing.
A-structure: All characteristics are observed.
T-structure: Only characteristics 1 and 4 are observed. (Note that characteristic 4 is identical to 3
for A structures.)

Figure 2: A-Structure Logistics Channel

Finished
Tractor
I
---------------------------------------------
/ I \
Purchased Transmission Engine
Components ________________ --------
/ I \ / \
Part Part Part Purchased Part
GMT Deere Deere Components GMT
I I I I
Op30 Op20 Op30 Op20
* MC 3 MC22 MC23 MC 5
I I I I
Op20 Op10 Op20 Op10
MC 5 MC21 MC22 * MC 3
I I I I
Op10 RM 3 Op10 RM 5
MC2 fdwy MC21 fdwy
I I
RM 2 RM 4
fdwy outside
fdwy

Op = Routing operation
MC = Machine center used
RM = Raw material type
* = Constraint
79 European Journal of Economics, Finance And Administrative Sciences - Issue 17 (2009)
As a result of this analysis it appears that TOC's V-A-T logical structure framework can be used
to describe an entire logistical channel rather than having to be applied to each separate organizational
unit.
Further, the resulting model appears to be more robust than the linear models by providing the
opportunity to uncover research questions concerning TOC's use in a supply chain. For example,
several components of TOC scheduling used in a production line can be identified and examined
within the context of a logistics supply chain. Thus, the following questions also emerge:
1. If a single constraint for the entire logistics system exists, how could it be identified across
organizational boundaries?
2. Can the overall throughput of the entire logistics system be improved by applying TOC's
drum-buffer-rope scheduling methods?
3. Can the continuous improvement methods used in TOC (the five-step process) allow for
improvements to be identified and focused regardless of organizational boundaries?
4. Can V-A-T analysis be used to identify critical control points within a supply chain to
facilitate effective supply chain management?
5. Can the use of TOC methods improve the overall throughput of the supply chain in a
migration to a full JIT environment?
6. What performance measurement system can be developed to allow supply chain partners to
adopt a global versus a local viewpoint?
7. Can all of the activities within a V-A-T logical supply chain structure be synchronized to the
rate of output at the constraint activity?
8. What approaches can be developed to identify the supply chain's overall limiting factor
(constraint)?
9. What are the methods that can be developed to facilitate the supply chain partners working to
improve the capability and the output of the entire supply chain?
These observations can be tested either by field experiments in the logistics channel itself, or by
using simulation methods. Further work in that regard is necessary. However, it appears, based on this
case, that Burnham's observations concerning the utility of viewing the entire logistics system as one
entity to manage and improve is correct. TOC may be the vehicle by which the benefits of JIT can be
fully attained across organizational boundaries.

References
1] Bowersox, D.J., Closs, D.J. and Helferich, O.K., (1986). Logistical Management, (3rd. ed),
New York, NY: Macmillan Publishing Company.
2] Bowersox, D.J., Daugherty, P.J., Droge, C.L., Germain, R.N., and Rogers, D.S., (1992).
Logistical Excellence, Digital Press, Burlington, MA: Digital Press.
3] Burnham, J.M., (1992)."Systematic Logistics Improvement: Integrating Principles and
Practices", 35th International Conference Proceedings, Falls Church, VA: American
Production and Inventory Control Society, 500-510.
4] Dobler, D. W. and Burt, D. N., (1996). Purchasing and Supply Management, NY: McGraw-
Hill Companies, Inc.
5] Goldratt, E. M., (1984). The Goal, Croton-on-Hudson, NY: North River Press.
6] Goldratt, E. M. and Fox, Robert E., (1986). The Race, Croton-on-Hudson, NY: North River
Press.
7] Goldratt, E. M., (1990). The Haystack Syndrome, Croton-on-Hudson, NY: North River Press.
8] Goldratt, E. M., (1992). The Goal (2nd. Revised Ed), Croton-on-Hudson, NY: North River
Press.
9] Handfield, R. B., & Nichols, E. L., (1999). Introduction to Supply Chain Management, Upper
Saddle River, NJ: Prentice-Hall.
80 European Journal of Economics, Finance And Administrative Sciences - Issue 17 (2009)
10] Lehmann, D.M., (1996). "Integrated Enterprise Management: A Look at the Functions, the
Enterprise, and the Environment", 39th International Conference Proceedings, Falls Church,
VA: American Production and Inventory Control Society, 104-107.
11] Lockamy, A. and Cox, J. F., (1991)."Using V-A-T Analysis for Determining the Priority and
Location of JIT Manufacturing Techniques", International Journal of Production Research, 29
(8), 1661-1672.
12] Marshall, C., and Rossman, G.B., (1990). Designing Qualitative Research, Newbury Park, CA:
Sage Publications, Inc., 320-321.
13] McMullen T.B.,(1996)."Theory of Constraints: The Infrastructure if Agile Manufacturing
Service", 39th International Conference Proceedings, Falls Church, VA: American Production
and Inventory Control Society, 336-346.
14] Minsky, M.J., (1996). "Direction of Supply Chain Optimization Technologies", 39th
International Conference Proceedings, Falls Church, VA: American Production and Inventory
Control Society, 201-205.
15] Mlot, B., DiFrancesco, L, Perry, D., Landvater, D. and Martin, A., (1996). "Distribution
resource planning: The Critical Link form Final Point of Manufacture to Final Point of Sale",
39th International Conference Proceedings, Falls Church, VA: American Production and
Inventory Control Society, 294-297.
16] Monczka, R., Trent, R. & Handfield, R. B., (1998). Purchasing and Supply Chain
Management, Cincinnati, OH: South-Western Publ.
17] Novack, R.A., Rinehart, L.M., and Fawcett, S.A., (1992). "Rethinking Integrated Concept
Foundations: A Just-in-Time Argument for Linking Production/Operations and Logistics
Management", International Journal of Operations & Production Management, 13 (6), 31-43.
18] Patton, M. Q. (1990). Qualitative Evaluation and Research Methods (2nd ed.), Newbury Park,
CA: Sage Publications, Inc.
19] Rosen, L.D.,(1996)."Service: The Next Frontier", 39th International Conference Proceedings,
Falls Church, VA: American Production and Inventory Control Society, 41-44.
20] Scully, J. and Fawcett, S.E., (1993)."Comparative Logistics and Production Costs for Global
Manufacturing Strategy", International Journal of Operations & Production Management, 13
(12), 62-78.
21] Spencer, M.S., (1993). "Warehouse Management Using V-A-T Logical Structure Analysis"
International Journal of Logistics Management 4 (1) pp. 35-47.
22] Spencer, M.S. and Cox, J. F., (1995)."Optimum Production Technology and the Theory of
Constraints: Analysis and Genealogy", International Journal of Production Research, 33 (6),
1495-1504.
23] Stock, J.R., and Lambert, D.M. (1997). Strategic Logistics Management (4th. ed), Homewood,
IL: Irwin.
24] Strauss, A. L. (1990). Qualitative Analysis for Social Scientists, Cambridge, England:
Cambridge University Press.
25] Toni, A.D., Filippini, R., and Forza, C., Scully, (1992) "Manufacturing Strategy in Global
Markets: An Operations Management Model", International Journal of Operations &
Production Management, 12 (4), 7-18.
26] Umble M., and Srikanth, M., (1990). Synchronous Manufacturing, Cincinnati, OH: South-
Western Publishing Co., 221-255.

View publication stats