Table of Contents
Go Back I. Martinez, G. Wadhwa, and R. MacDonald
Evolution of a System Dynamics Intervention:
How changing the rules in a small health care private practice can redefine the strategic position
of the firm and increase overall performance.
Ignacio J. Martinez-Moyano' Gary Wadhwa Roderick H. MacDonald
Doctoral Student Maxilofacial Surgeon President
University at Albany Adirondack Oral & MindWalk Consulting
Maxillofacial Surgery
im7797@albany.edu awadhwal@nycap.rr.com rod@mindwalkconsulting.com
(518) 442-5257 (518) 424-2940 (518) 458-7976
Abstract
This paper reports an ongoing project using system dynamics modeling as the unifying framework for
understanding how to change and improve the way a small health care practice is managed. Through the
development of the project we have used group model building sessions, one-to-one exploration of structural
explanations, and extensive model building and testing to clarify hypotheses related to different areas of the practice
considered key by our clients. Major insights found include: strategic management of accounts receivable, a switch
in strategic orientation of the practice and its implications, and the realization of backlog of patients as a key driver
of the firm’ dynamics. Based on our experience, a general framework for system dynamics interventions is
presented. Additionally, three system dynamics models developed for the study are presented and explained.
Keywords: Health-Care Dynamics, System Dynamics, Innovation, Implementation, Best Practices, Rules, Rule
Making, Rule Dynamics
' Corresponding author at im7797@albany.edu
I. Martinez, G. Wadhwa, and R. MacDonald
Introduction
The Phases of the Intervention Process
Process and Products...........0000
First Phase—Initiation—Getting acquainted with system dynamics
The Concept Model Used..........
Second Phase—Organizational Learning
The Key Reference Modes.
Third Phase— Diffusion—Extensive use of system dynamics as guiding strategic and
innovative framework
Operations Sector
Financial Sector ..
Fourth Phase—Consolidation—Introduction of system dynam: is standard operating
procedure—rule—in management practice in the firm.
Final. COMMENES' ssissvsiessscvsvsevsesevsrsessvessssrveveavensaseseets
References.........ccceee
(1) Health-Care 1 Mode
(2) Operations 1 Model.
@) Finaticial 1 Modelisssscnniinannnnanumnnmaar aeRO NE
I. Martinez, G. Wadhwa, and R. MacDonald
Introduction
This paper presents a case study of a system dynamics intervention undertaken by a small private-
practice health care provider in upstate New York specializing in oral-maxilofacial surgery. The purpose
of this intervention was to improve the financial performance of the practice and to understand how best
to elicit and provide system dynamics insights into a small for-profit organization.
The practice was established at one location in 1994 and expanded to a second location in 1999.
The practice had financial growth of 1250% from 1994 to 2001 and grew from 4 to 26 employees in the
same period of time. By December of 2002, the practice offered this array of services: oral surgery
implants, oral pathology, and facial cosmetics in their office practice, and they provided hospital-based
facial trauma, tumors, and reconstructive surgery. In late 2001, the practice began to experience reduced
earnings. Moreover, concerns about quality control and staff turnover began to increase. The managing
partner was concerned about this and hoped to head off problems before they became acute. To
accomplish this, the managing partner began identifying “best practices” inside and outside of the
industry (six sigma quality control, lean manufacturing principles, etc). These best practices were
implemented to improve and standardize both quality and operational capabilities. These innovations
proved efficient and effective (for examples of innovation implementation in different industries see
Cobbenhagen, 2000), yet the managing partner felt that the hoped for results were never realized.
The managing partner’s search for improved ways of managing growth lead him to system
dynamics modeling and systems thinking tools. The concept of feedback processes generating behavior
(Forrester, 1975; Richardson, 1991; Sterman, 2000) resonated well with the managing partner.
Furthermore, the managing partner’s philosophy has always been that he needed to be capable of
understanding and implementing the management tools necessary for operating a successful practice. In
addition, he felt that his staff had to participate in the learning process in order to understand and accept
the changes he was asking them to make.
I. Martinez, G. Wadhwa, and R. MacDonald
In order to bring himself up to speed, the managing partner took formal classes in system
dynamics modeling and brought in consultants to assist with group model building (to bring the staff
along), to assist with the development of formal and informal system dynamics models, and to act as a
system dynamics coach so that the managing partner could begin to develop his own models of important
organizational issues. The intervention encompassed multiple phases that involved group model building,
coaching, and the development of formal models.
The Phases of the Intervention Process
The intervention process was, after the fact, divided into four distinct phases that served different
purposes, activities, time frames, and resulted in various outcomes. The involvement of the managing
partner of the practice has been key in the development of the project. The managing partner has acted as
client, group participant, gatekeeper for the group model building process (Andersen and Richardson,
1997; Andersen, Richardson and Vennix, 1997), project champion and a member of the modeling team.
The project phases are:
. First Phase—Initiation
o Purpose: Becoming acquainted with system dynamics (managing partner and staff).
o Main Activity Developed: Taking formal courses in systems thinking and system
dynamics to enhance modeling skills (managing partner).
co People Involved: Managing partner and staff.
o Time Frame: 12 months.
o Main Outcome: Identifying the powerful role system dynamics can have in managing the
firm.
. Second Phase—Organizational Learning
co Purpose: Introducing system dynamics to the firm.
o Main Activity Developed: Group sessions to explain system dynamics; group sessions for
developing concept models; individual sessions expanding on insights generated by
modeling; and identifying leverage points in the system
o People Involved: All employees.
I. Martinez, G. Wadhwa, and R. MacDonald
Time Frame: This phase lasted 12 months; however, the actual time staff was involved
varied. Group sessions were scheduled for every other month. In addition, during
weekly group sessions some aspects of the system dynamics based interventions were
explored and explained in detail.
Main Outcome: All employees were introduced to system dynamics terminology and
icons and the purposes for using system dynamics models. Health-Care 1, a system
dynamics model, was developed (reported on Martinez-Moyano and Wadhwa, 2002).
Third Phase—Diffusion
Purpose: Use of system dynamics to guide strategic decisions.
Main Activity Developed: Introduction of system dynamics models developed from
weekly staff meetings.
People Involved: All staff members, modelers, and the occasional facilitator.
Time Frame: Six to nine months.
Main Outcome: Organizational recognition of the ‘new’ way of identifying ‘what is
happening and why’. New and improved understanding of how my work influences
performance of the firm (why my work is relevant).
e Fourth Phase—Consolidation
°
Purpose: Introduce system dynamics as part of the standard operating decisions for
management practice in the firm.
Main Activity Planned: Change in standard formal procedures to include system
dynamics based elements, more formal training in system dynamics for the staff
members, and establishment of strategic alliance with modeling experts to supervise the
in-house work.
People Involved: Managing partner, modeling coaches, and facilitators.
Time Frame (planned): Six to twelve months.
Main Outcome (expected): Deeper understanding of how to include dynamic
considerations in management, change in culture in the organization to adopt a new
‘norm’ that includes dynamic thinking and holistic approaches to management problems,
and linking the new way of looking at the firm’s problems and opportunities with the
day-to-day decision making processes.
I. Martinez, G. Wadhwa, and R. MacDonald
The timeline for the intervention is shown in Figure |—above. We present the general
timeline of activities to show how the different phases evolve over time.
2000 Year 1 2001 Year 2 2002 3Year 3 2003 Year 4
1 2 3 4 1 2 3 4 12 3 4 123 4
Phase 1
Initiation
Formal Instruction in SD
Modeling exercises
Development of concept
Phase 2
Organizational Learning
Learning SD group session
Group Model Building
Identification of insights
Where do | influence the
Phase 3
Diffusion
Introduction of elements in
staff meetings
Definition of new projects
within the firm
Clarification of the relevance
of my work
Phase 4
Consolidation
Change of formal procedures
to include SD-based elements
Formal Instruction in SD
Strategic alliances
Figure 1—Timeline
Process and Products
First Phase—Initiation—Getting acquainted with system dynamics
The first phase was structured with the goal of having the client group participate in the
development of the formal model in order to foster model ownership. To achieve this goal while
simultaneously working on research through ‘real action’ in the organization (for action research elements
see Argyris, Putnam and Smith, 1985; Argyris and Schén, 1996) one-hour weekly meetings, with the
practice’s management team, were undertaken. These meeting lasted two months and the group decided
that the products the modeling effort would attempt to generate were:
I. Martinez, G. Wadhwa, and R. MacDonald
1. Structural understanding of the elements that causes the behavior observed of the Health-
Care Practice.
2. Dynamic understanding of the practice.
3. A policy testing instrument to enhance the practice’s strategic planning capabilities.
4. A means to increment the development of the firm’s organizational intelligence.
The Concept Model Used
During the initial part of the intervention (following Andersen and Richardson, 1997; and
Andersen, Richardson and Vennix, 1997) a concept model was used. The concept model used (see Figure
2, below) was designed to capture the staff’s attention to the dynamic process of the firm and to convey
the stock and flow terminology.
Cash at hand
Cash In Cash out
Potential Investment
Income
in Capacity + Variable operating
Total Number of costs
Procedures/quarter
Capacity Unsatisfied
Unhappy patients =
Patients NY Total Number of —
Patients Served word of mouth
incoming patients bad word of
Satisfied sina
‘Completion of | |_Patients
Patients
$e} ies Trust
Net Increase in ctors ‘Net Trust Change
Patients per unitof — Referal Base Se
“oupaciy Trust Normal ‘Time for trust to
va breakdown
__ Effect of Trust
Figure 2—Concept Model
Second Phase—Organizational Learning
Introducing System Dynamics to the Firm
The Key Reference Mode
To understand the system through a group model building approach, one key element is to elicit
the variables and reference modes of importance, to the group (Andersen and Richardson, 1997), in an
I. Martinez, G. Wadhwa, and R. MacDonald
effort to capture their view of the system. The group identified the following four variables as being the
most relevant for this study. Those four variables are: (1) employee workload, (2) employee productivity,
(3) perceived quality, and (4) net earnings. The group created reference modes for these variables
expressing desires and fears about what would or could occur in the future. Figures 3 to 6 show the
reference modes for the variables. The desired behavior is identified with a dashed line and the feared
with a solid line.
Employee
Employee Work Load Productivity
1995 ——+» 2010 1995 ——» 2010 1995 ——> 2010 1995 ——+» 2010
Figure 3—Employee Figure 4—Employee — Figure 5—Perceived Figure 6—Net
Work Load Productivity Quality Earnings
According to the group, workload had been increasing while productivity had stalled. This
influenced the perception of quality and results in stagnant earnings. With these four variables in mind
and relying on the concept model as an elicitation device, we worked with the group to create a feedback-
centered understanding of the dynamics of the organization. The group identified eight major causal
loops; two of them are shown in Figures 7 and 8.
I. Martinez, G. Wadhwa, and R. MacDonald
The Causal Loops
Effect of Word of
Mouth on Referrals +
Referrals
RI
Word of Mouth
4
Treated patients
New aa
Backlog of
“Se patients to be
treated
+ el
i) " S
* of ws Work Clo
Manta o NL &
Figure 7—Workload Loop
In the firm, the workload loop, Figure 7, is a central loop for explaining pressures on growth
processes and the dynamic behavior observed by participants and shown in Graphs 3 to 6. The group
believed that workload influences productivity and the quality of services as well as being a key element
of stress generation.
Figure 8 shows the productivity/staff loop that allows us to clarify the effect that staff have on
productivity and the way in which the different pressures are generated. As the number of treated patients
grows, available income increases, influencing the ability to hire new staff and through an augmented
total staff, influence the workload. This is positive in the sense that staff will have enough time to deliver
quality services and stress levels could be reduced. However, additional staff requires that experienced
staff be allocated time to train the new staff in the specific processes of the firm. General knowledge is
obtained by new staff from education and experience in other practices, but specific knowledge related tot
he firm must be provided by experienced individuals within the firm (Brickley, Smith and Zimmerman,
2001, pp. 341-342).
I. Martinez, G. Wadhwa, and R. MacDonald
Effect of Word of,
Mouth on Referrals +
Word of Mouth
bi +
Referrals.
RI
Treated patients
a 9
+
New an Income
Backlog of
~N patients to be
treated (wf
lo 4 Productivity = +
ae Ability to hire
°. of Services Work Load:
New Staff
~—AZ Veas As)
du” > Stress
Time to change
inexperienced to
Experieced A R4
Total Staf
Experienced staff
devoted to Training
Figure 8—Productivity/Staff Loop
The Health-Care 1 System Dynamics Model
The original model had been conceptualized in five Sectors (Operations, Community,
Knowledge-Based Innovation Projects, Human Factor, and Financial). Figure 9 shows the sector diagram.
For the development of understanding of how the practice was managed, identifying the different
sectors was key. In the beginning of this intervention process, in our conversations, the financial sector
kept on being described as the most important one in the practice. Over time, the conversations went to
the other sectors as the realization of the interconnectedness of elements in the firm arose. Members of the
firm recognized this insight as very important because it allowed them to ‘connect’ their individual
activities to the bottom line of the practice in a very clear and simple way. The development and use of
the sector diagram became one appealing tool to communicate the system’s structure to the staff.
10
I. Martinez, G. Wadhwa, and R. MacDonald
Operations Community
Satisfactory Service
© Administrative = | ----------------- >
and operational New Knowl * Patients
senders ld aiciarnicy oe oe = Families, relatives,
+ Capacity nese hanes i and friends
+ Quality of ewe nsigl + = Referring Doctors
Service ‘= Insurance Companies
6 eeiace Git Knowledge-Based * Professional Assoc
Innovation Projects || ;,,cerutstionand
1 x ,
a j © Change Programs Git, Renee:
+ Administrative improvement x 1
Human Factor + Doctor's involvement G v
+ Intensity of projects : :
+ Administrative staff + Inbdiigence poveration Financial
+ Doctors ere
+ Skills & Knowledge T ry ry = Cash flow
+ Stress & Synergy} 1 1 «Investments
Motivation i ' '
+ Training Capacity | Funding! * Profitability
» Work boad end pi ay ‘+ Sustainability
Productivity
* Culture Recognition of Effort
Figure 9—Sector Diagram
The Behavior of the Model
The modeling effort was conducted following standard practice to enhance quality and
confidence in model results (for an explanation of the process followed see Richardson and Pugh, 1981;
Martinez and Richardson, 2001; and Martinez-Moyano and Richardson, 2002). Figures 10 and 11 show
the base-case behavior and the improved-case behavior of the model.
The base run considers a 20% increase in the average tasks per patient. This run is considered the
base-case run because due to the implementation of innovations and increased administrative controls the
average number of tasks that the firm has to perform had incremented. Innovation and uncertainty go
hand in hand; successfully implementing innovations require a different management style and
organization than the one used in steady-state processes (Cobbenhagen, 2000, p. 277) . Companies should
recognize this and change the way they conduct business during the process—steady-state equilibrium,
transient-state dynamics, and new equilibrium of the firm.
11
I. Martinez, G. Wadhwa, and R. MacDonald
Key Variables
1.2 Dmal
1.05 Quality
40,000 Dollars/Month
135. Tasks/(Month*Staff)
0.9 Dmnl
0.9. Quality
-2,000 Dollars/Month
90. Tasks/(Month*Staff)
0 2 4 6 8 10 12 14 16 18 20 22 24
Time (Month)
WorkLoad Ratio : Base Run Dmal
Perception of Quality : Base Run Quality
Net Income : Base Run Dollars/Month
Productivity : Base Run “ * Tasks/(Month*Stafi)
Figure 10—Base-case Run
Key Variables
1.2 Dmal
1.05 Quality
200,000 Dollars/Month
135. Tasks/(Month*Staff)
0.8 Dmnl
0.9. Quality
0. Dollars/Month
90. Tasks/(Month*Staf)
0 2 4 6 8 10 12 14 16 18 20 22 24
Time (Month)
WorkLoad Ratio : Improved Run Dmnl
Perception of Quality : Improved Run Quality
Net Income : Improved Run Dollars/Month
Productivity : Improved Run. #444 Tasks/(Month*Staft)
Figure 11—Improved-case Run
In order to identify successful innovations, we need to know what does it mean in the context of
the firm. Cobbenhagen (2000, p. 71) offers a definition of successful innovation as being ‘the economic
exploitation of innovation’ and saying that it is difficult to identify a way to understand the successfulness
of the innovation in firms. Based on the results of the study conducted, we say that successful innovations
12
I. Martinez, G. Wadhwa, and R. MacDonald
are those that can act as levers for the attainment of goals that are dynamically coherent and systemically
desirable (Lane and Oliva, 1994) in the context of a culturally-feasible change.
According to Cobbenhagen (2000, p. 273) three elements are necessary to create successful
innovations: a strong knowledge base, ability to proactively manage innovations, and ability to manage
the relationship with the environment. The improved-case run, shown in Figure 11, uses these concepts to
create better behavior of the system. In this run, a 20 percent decrease in the average tasks per patient is
simulated along with doubling the average number of contacts per referring doctor from 11 to 22. These
changes assume that the innovation is proactively managed (the application of the lean concepts to
decrease the number of tasks), that a strong knowledge base is created (to actually induce the changes),
and that the relation with the environment is managed adequately (by means of increased contacts with
refereeing doctors). This improved run is just a first approximation to a more complete exploration of the
complex set of combinations and possibilities present in this case study. It is now clear that innovation
management comprises many ingredients from complexity—large number of variables involved, tightly
interrelated in non-linear fashion, and highly dynamic—that makes it both an interesting research theme
and particularly suited for being studied using system dynamics (Milling, 2002, p. 85).
Third Phase— Diffusion—Extensive use of system dynamics as guiding strategic and innovative
framework
The sectors further developed presently in the firm include:
1 Operations Sector
a. Capacity
2. Financial Sector
a. Cash Flow/Accounts Receivable
13
I. Martinez, G. Wadhwa, and R. MacDonald
Operations Sector
The diagram presented in Figure 12 captures the primary feedback loops observed in the larger
formal model of the operations sector of the practice. The practice had historically considered and still
considers quality to be of the utmost importance. Allowing quality to slip in order to see more patients or
to reduce waiting time for routine procedures was never considered acceptable.
The practice had performed six-sigma studies that had worked to standardize routines in order to
reduce errors, replication of work, and to increase efficiency and productivity. What the model indicated
was that the practice should focus on the backlog of patients waiting for routine services. The initial
group model-building project resulted in the identification of workload as being at the center of numerous
feedback loops. The formal model also identified the workload as being critical, but with limited
technological improvements being available to increase productivity. The implementing of six-sigma
standards having already occurred in the practice and quality in a healthcare practice being relatively
fixed by standard operating procedures, left management as one of the few places in this system that had
control over limiting the backlog of patients waiting for services.
For the practice in this case, being proactive and limiting the growth of the backlog of patients by
means of refusing to accept new referrals had not occurred before. Traditionally, all referrals and all
referring doctors had been accepted (after all, getting more business in is always better according to
normal intuitive thinking). Patients with acute problems would be scheduled quickly, but patients without
acute problems were given the next available date for treatment. Under these conditions patients who felt
that the wait was too long would go elsewhere. Furthermore, referring doctors who felt that their patients
were waiting to long for appointments would also begin to refer a larger portion of their patients to other
specialists.
14
I. Martinez, G. Wadhwa, and R. MacDonald
; «————— Wordof
Effect of Word of “+ Mouth
Mouth on Referrals
Effect ofthe |“ __~Delay in Receiving
Treatment Delay on Treatment
Referrals B4 i 4 ae
m| Backlog of |g Treated
__p New Patients +] Patients Patients
on i Apt) eo
| 4
Workload + Productivity
Quality of 3 —~
ra
Ability to Hire
Total Staff New Staff
a
Figure 12—Operations Sector
fie
Stress
ee
When the model revealed this behavior the managing partner noted that historically there had
been an oscillation in referring doctors and in the waiting time for treatment. This oscillation had always
been attributed to the economy, relationships with referring doctors, and changing health care coverage
policies. The model indicated that those exogenous forces might not be solely responsible for the
oscillation in the number of patients the practice treated. The idea that waiting time for treatment was
important to both patients and referring doctors resonated well with the managing partner. Furthermore,
the model revealed that a proactive approach, where policies to weed out referring doctors who did not
refer a profitable and interesting mix of patients, could be implemented when the backlog got high.
The practice began to keep a database of referring doctors and patients. When the backlog got
too large the number of referring doctors was selectively cut. Based on a set of criteria the practice was
able to decide whom they would prefer to work with. This resulted in a reduction in staff turnover, the
ability to see patients quickly, the ability to spend additional time with patients and thus generate a
15
I. Martinez, G. Wadhwa, and R. MacDonald
perception of quality, and earnings increased as referring doctors dumping less profitable patients on the
practice were eliminated. For the managing partner of the healthcare practice, the realization that the
practice had more control over what was happening than originally thought was enlightening.
Financial Sector
The structure shown in Figure 13 captures the aging chain for late payments. The practice had
resources allocated along the aging chain to capture funds that were owed, but not yet paid. Any errors in
coding resulted in the rejection of claims by third party payers (insurance companies). These errors
would need to be identified, corrected and the bills resubmitted. The third party payers are under no legal
obligation to make payments on bills submitted correctly after 90 days of the treatment rendered. The
practice must then petition the insurer or request the patient to pay the bill out of pocket. This results in
reduced cash flow, due to waiting for payment, and loss of income as only a portion of the late bills tends
to be paid after too much time has elapsed.
The structure in Figure 13 was developed with the senior partner to capture the aging chain for
late payments. From this point it was anticipated that this structure would be elaborated and the
resources allocated and the decision rules for allocating those resources would be captured. However,
from this structure the senior partner realized that the policies they had been using of rewarding
employees working on late payment accounts based on the amount of collections they made was not the
best policy. The senior partner shifted internal personnel from collections to initial billing. Training was
conducted on proper billing procedures for all staff members and the incentive policy was changed from
one that focused on collections to one that focused on the reduction of billing errors. The pool of late
payments that had accumulated was handed off to resources outside of the practice. Although the
decision to reallocate resources and change the incentive policy was based on a partially developed
model, the decision proved to be correct in that the practice’s cash flow was noticeably increased and the
number of bills that turned into late payments was significantly reduced.
16
I. Martinez, G. Wadhwa, and R. MacDonald
Fourth Phase—Consolidation—Introduction of system dynamics as standard operating
procedure—rule—in management practice in the firm.
In this phase, system dynamics is intended to become a standard operating procedure for
management in the organization. Two examples of the types of results that the intervention has generated
are new momentum policies being generated and additional recognized insights that have guided new
decisions and actions in the firm.
The Momentum Policies Generated
The group, analyzing the simulation results from the different models, proposed several policies.
The policies are:
1. Training programs for all levels of employees
Cross training in multiple skills
Standardizing of processes
| ee ad
Slowing down the pace of the practice
The Recognized Insights and Recommended Policies
The modeling process allowed the group to recognize insights about the practice. The insights
belong to three ‘major’ areas. The areas are:
e Main Drivers of the Dynamics of the Practice
o Backlog of patients was recognized as a key-leading indicator of the way the practice was
performing. By concentrating in monitoring the way in which the backlog and the
average time to be served behaved, the practice can take dynamic adjustments for
improved performance.
o The importance of the word-of-mouth effect in the firm’s behavior was identified as
another key driver of the dynamics. One participant said: “after knowing all of this, you
just cannot concentrate on the fee schedule any more [as we did before]. You have to pay
attention to the dynamics involved.”
17
I. Martinez, G. Wadhwa, and R. MacDonald
INew Bills
“AIR Greater
gee than 120>
Total Late Accounts
Receivable
ae
“Recounts “Rewourks Aone Say
au ai ger! Receivable (61 BH) Receivable (20
10 30) ‘Aik Aging 1 10.60) 1090) AiR Aging 3 10 120) AVR Aging 4
Outflow from AR Outflow for Ak 90 Oulow for A/R 120
‘Outflow from 60
AIR Paid for 0t0 ‘AIR 30
ODay Paid for 3110 gh
sg Aik Pade 31 an Pai SO Dy ‘AIR Paid for 120
on <30 bays Day
<30 Days
30 Days
Fraction Paid
AR 30
<AIR Aging 4>
om
Fraction Paid
AUR 60
New Bills Not Paid
in 120 Days
Fraction Paid
AIR 90
Time to Write off A/R
AVR Greater
than 120
AR Waiten off
<30 Days
Fraction Paid
AIR 120
Figure 13—Part of the Structure of the Financial Sector
°
I. Martinez, G. Wadhwa, and R. MacDonald
Strategic Orientation of the Practice
Focusing on implants and recognizing the borders of their practice as ‘extended’ towards
the referring doctors, crown technicians, and color specialists is identified as a key
strategic position for the practice. Having the orientation towards implants attract more
and higher-end patients that allow the practice to maintain quality levels and revenue
streams. However, this new orientation is challenging because the integrated quality of
the implant, as seen by the patient, includes operations outside of their ‘traditional’
control. The interorganizational relationships present in their practice were identified as
critical to be able to control the operations.
e Innovations in Management
°
Innovations in management practices will be adopted as a rule to change existing
practices such as in the case of accounts receivable. This process will be incorporated
into the day-to-day analysis of operations using a system-dynamics based framework.
The Recognized Insights about how to use System Dynamics in the Firm
Besides the insights pertinent to the firm’s operations, the group was able to identify some
interesting insights about the use of system dynamics in the organization. In this part of the reflection, the
facilitator and the modeler were very active trying to help the larger group clarify these ‘nuggets’ of
knowledge about the intervention process. At this point, many individuals in the firm were very interested
not only in what was happening but in how it was happening as well. The insights are organized in four
categories:
e Types of Models
°
Trying to generate and present to the group the simplest model possible that captures the
dynamics under study was considered key. This simplicity was considered very important
to be able to have people relate to the model and its possible lessons. This is consistent
with ideas related to ‘insightful little models’ (Richardson, 2000) for enhanced
I. Martinez, G. Wadhwa, and R. MacDonald
understanding of dynamic phenomena. One participant expressed this by saying: “this
should not be a Ph.D. exercise, it should be something that people can relate to.”
You have to clarify from the very beginning with everyone that, as Sterman (2002)
mentions, all models are wrong and all models are limited representations of reality. This
initial recognition clears the path for increased participation of the people in the firm
because the level of anxiety about requiring a good model goes down and their creativity
goes up because they realize that there is always room for improvement in the models
that they are working with.
e Types of Variables
°
The types of variables that should be used are those that have an appeal to the individuals
in the organization. Being able to ‘talk the walk’ of the firm increases the possibility of
integrating the new ideas to the firm’s dominant cultural stream.
e Types of Interaction
°
The interaction between the ‘external’ and the ‘internal’ people involved in the process
should be as simple as possible, or at least it should be presented as simple. The use of
simple tools, simple concepts, and simple mechanisms can be very powerful for the firm
and can enhance the level of collaboration to make it a real solution and a new way of
doing things in the organization: a new norm. These ideas are consistent with ideas
relating the importance of managing the interface between the modeling group and the
client group in relation the expectations of both parties (Andersen and Richardson, 1997;
Haslett, 2001).
e = Types of Results
°
All the dynamics that you see in large corporations evolved here. Even though this case
was developed in a small health care practice with a very high degree of centralization
and a high capacity to command people to do it, policy resistance and unanticipated
consequences arose. The group had to internalize sufficient knowledge and confidence
about the process to be a success. This confirmed us that a system dynamics intervention
cannot be mandated without having unanticipated consequences that can defeat the
intervention. The autonomous evolution generated in system dynamics interventions
needs to be recognized and managed in order to be able to change the firm’s culture.
20
I. Martinez, G. Wadhwa, and R. MacDonald
Final Comments
This paper has presented an explanation of our experience in an ongoing organizational
intervention using system dynamics modeling as the framework. This process has been both rewarding
and intriguing for us. Some interesting questions have arisen from this experience. We have been trying to
think if the way we have conduced this intervention is something that other practitioners could use to
improve their practice. Additionally, we have identified some characteristics about our wok that make it
desirable. These characteristics are:
1. This type of intervention seems to generate higher levels of stickiness of the results and
benefits in the organization.
2. This intervention generates ‘shared realities’ for the members of the organization to
consider.
3. This type of intervention tends to infuse system dynamics ideas in the day-to-day
activities of the members of the organization, becoming a vehicle for cultural change
affecting the ‘norms’ of the organization.
4. This type of intervention generates a new language that enables both a new type of
dialogue in the organization and the necessary process for it to become effective. This
new language and dialogue allows for the current culture to evolve towards a new culture
in the organization. One very important product that system dynamics interventions
generate is the creation of a new way of looking at the world and a new way to express
what we see in the world. This should not be seen only as a ‘by-product of our work’ as
Campbell (2001, p. 210) describes. In our interventions, it should be an important and
desirable main product and contribution for change in the firm (for further reading about
language, dialogue, and the way it influences groups and organizations see Senge, 1990;
Bohm and Nichol, 1996; Isaacs, 1999).
5. New rules seem to evolve from the intervention as a natural process without having to be
formalized from the beginning.
21
I. Martinez, G. Wadhwa, and R. MacDonald
However, the disadvantages of this type of intervention seem to be:
1. This type of intervention seems to be very time consuming and potentially long (is this
bad?).
2. This type of intervention generates a new power within the organization that can be
‘misused’ creating a ‘new’ culture that is not as effective as the one that was originally in
place.
3. This type of intervention can be costly in time and resources.
4. This type of intervention seems to be very sensible to having a strong internal champion
to work with. If you do not have a very committed internal ‘champion’ in the
organizations the probability of success seems low.
For certain, additional research appears necessary.
22
I. Martinez, G. Wadhwa, and R. MacDonald
References
Andersen, David F. and George P. Richardson (1997). "Scripts for Group Model Building." System
Dynamics Review 13(2): 107-129.
Andersen, David F., George P. Richardson and Jac A.M. Vennix (1997). "Group Model Building: Adding
More Science to the Craft." System Dynamics Review 13(2): 187-201.
Argyris, Chris and Donald A. Schén (1996). Organizational Learning II: Theory, Method, and Practice.
New York, Addison-Wesley Publishing Company.
Argyris, Christopher, Robert Putnam and Diana McLain Smith (1985). Action Science: concepts,
methods and skills for research and intervention. San Francisco, California, Jossey-Bass.
Bohm, David and Lee Nichol (1996). On dialogue. London ; New York, Routledge.
Brickley, James A., Clifford W. Smith and Jerold L. Zimmerman (2001). Managerial Economics and
Organizational Architecture. Boston, MA, McGraw-Hill Irwin.
Campbell, Deborah (2001). "The long and winding (and frequently bumpy) road to successful client
engagement: one team's journey." System Dynamics Review 17(3): 195-215.
Cobbenhagen, Jan (2000). Successful Innovation. Cheltenham, UK, Edward Elgar.
Forrester, Jay Wright (1975). Counterintuitive Behavior of Social Systems. Collected Papers of Jay W.
Forrester. Cambridge MA, Productivity Press: 211-244.
Haslett, Tim (2001). "Experiences and reflections on the transition from classroom to practice." System
Dynamics Review 17(2): 161-169.
Isaacs, William (1999). Dialogue and the art of thinking together. New York, Currency Doubleday.
Lane, David C. and Rogelio Oliva (1994). The Greater Whole: Toward a Synthesis of SD and SSM. 1994
International System Dynamics Conference, Sterling, Scotland, System Dynamics Society.
Martinez, Ignacio J. and George P. Richardson (2001). Best Practices in System Dynamics Modeling.
Proceedings of the 19th International Conference of the System Dynamics Society, Atlanta, GA
USA, System Dynamics Society.
Martinez-Moyano, Ignacio J. and George P. Richardson (2002). An Expert View of the System Dynamics
Modeling Process: Concurrences and Divergences Searching for Best Practices in System
Dynamics Modeling. Proceedings of the 20th International Conference of the System Dynamics
Society, Palermo. Italy.
Martinez-Moyano, Ignacio J. and Gary Wadhwa (2002). Modeling the Impact of Knowledge-Based
Innovations: The Case of Best Practices Implementation in a Small Health Care Private Practice.
Proceedings of the 20th International Conference of the System Dynamics Society, Palermo.
Italy.
Milling, Peter M. (2002). "Understanding and Managing Innovation Procesess." System Dynamics
Review 18(1): 73-86.
23
I. Martinez, G. Wadhwa, and R. MacDonald
Richardson, George P (1991). Feedback Thought in Social Science and Systems Theory. Waltham, MA,
Pegasus Communications.
Richardson, George P. (2000). Insightful Little Models. Systems Thinking and Dynamic Modeling: a
Conference for K-12 Education, Scamania Lodge, Washington.
Richardson, George P. and Alexander L. Pugh, III (1981). Introduction to System Dynamics Modeling
with DYNAMO. Cambridge MA, Productivity Press.
Senge, Peter M. (1990). The Fifth Discipline: the Art and Practice of the Learning Organization. New
York, Doubleday/Currency.
Sterman, John D. (2000). Business Dynamics: Systems Thinking and Modeling for a Complex World.
Boston MA, Irwin McGraw-Hill.
Sterman, John D. (2002). "All models are wrong: reflections on becoming a systems scientist." System
Dynamic Review 18(4): 501-531.
24
I. Martinez, G. Wadhwa, and R. MacDonald
Equations of the Models
(1) Health-Care I Model
JAG Ee nonaaeiodeieickickiccicck
Vil
Jnbodbeodobbinc obs ibbck
(002) "$/Patient High"=900
Units: Dollars/Patients
(003) —"$/patient Low"=350
Units: Dollars/Patients
(004) —"$/Patient Medium"=550
Units: Dollars/Patients
(005) Additions to Cumulative Patients=Outflow rate
Units: Patients/Month
(006) Adjustment time for moving staff=2
Units: Month
(007) Adjustment Time for Perception of Quality=1
Units: Month
(008) Adjustment Time for Quarterly Profits=6
Units: Month
(009) Average Out patient per group=Outflow rate
Units: Patients/Month
(010) "Average patients referred/month/referring people"=3
Units: Patients/Month/Referring people
(011) Average Quarterly Profits= INTEG (Net Change in Quarterly Profits, Estimated Quarterly Profits)
Units: Dollars/Quarter
(012) "Average Tasks/patient"="Average Tasks/patient Normal"*(1+STEP(Step in tasks, Time to step in tasks))
Units: Tasks/Patients
(013) "Average Tasks/patient Normal"=10
Units: Tasks/Patients
(014) "Average time to get experience/staff"=1
Units: Month
(015) Average Training per employee=Level Of Training/Total Staff
Units: Learning/Staff
(016) "Avg. Backlog to tasks"=Backlog of Patients/Tasks
Units: Patients/Tasks
(017) "Avg. Salary per staff per month"=4000
Units: Dollars/Month/Staff
(018) Backlog of Patients= INTEG (Inflow rate-Outflow rate,330)
Units: Patients
(019) Completion rate=Productivity*Total Productive Staff
Units: Tasks/Month
(020) Cumulative Patients= INTEG (Additions to Cumulative Patients,0)
Units: Patients
(021) Current Contact Per Month="Normal Contacts with People/Month"*Effect of the Workload Ratio on
Marketing
Units: Referring people/Month
(022) Current Fraction of High patients=Desired Fraction of High patients/Total Fraction
Units: Dmnl
(023) | Current Fraction of Low patients=Desired Fraction of Low patients/Total Fraction
Units: Dmnl
(024) Current Fraction of Medium patients=Desired Fraction of Medium patients/Total Fraction
Units: Dmnl
(025) Current Quality of Services=Effect of WorkLoad Ratio on Quality*Quality of Services normal
25
(026)
(027)
(028)
(029)
(030)
(031)
(032)
(033)
(034)
(035)
(036)
(037)
(038)
(039)
(040)
(041)
(042)
(043)
(044)
(045)
(046)
(047)
I. Martinez, G. Wadhwa, and R. MacDonald
Units: Quality
Desired Fraction of High patients=1/3
Units: Dmnl
Desired Fraction of Low patients=1/3
Units: Dmnl
Desired Fraction of Medium patients=1/3
Units: Dmnl
Desired Staff=28
Units: Staff
Desired training level=30
Units: Learning/Staff
Effect of Reputation on Loss of referral=f Effect of Reputation on Loss of referral(Reputation)
Units: Dmnl
Effect of Reputation on new referrals=f Effect of Reputation on new referrals(Reputation)
Units: Dmnl
Effect of the Workload Ratio on Marketing=f Effect of the Workload Ratio on Marketing(WorkLoad
Ratio)
Units: Dmnl
Effect of WorkLoad onProductivity=f Effect of WorkLoad onProductivity(WorkLoad Ratio)
Units: Dmnl
Effect of WorkLoad Ratio on Quality=f Effect of WorkLoad Ratio on Quality(WorkLoad Ratio)
Units: Dmnl
Estimated Quarterly Profits=Months Per Quarter*Net Income
Units: Dollars/Quarter
Experienced Staff= INTEG ((Gaining experience-Quitting)-Staff moving to training functions,28)
Units: Staff
"Experienced Staff (Training)"= INTEG (Staff moving to training functions,0)
Units: Staff
f Effect of Reputation on Loss of referral([(0,0),
(2,2)],(0,2),(0.293578, 1.8421 1),(0.617737,1.75439),(0.776758,1.68421
),(0.954128,1.47368),(1,1),(1.11927,0.631579),(1.22936,0.45614),(1.45566,0.254386
),(1.68196,0.166667),(2,0.1))
Units: Dmnl
f Effect of Reputation on new referrals([(0,0)-
(2,2)],(0,0),(0.238532,0.45614),(0.605505,0.807018),(1,1),(1.33945,1.2807),(2,1.5))
Units: Dmnl
f Effect of the Workload Ratio on Marketing([(0,0)-
(2,40)],(0,30),(0.0733945,29.4737),(0.140673,27.7193),(0.238532,25.4386),(0.35474,21.5789),(0.593272,1
0.513),(0.850153,1.10526),(1,1),(2,1))
Units: Dmnl
f Effect of WorkLoad onProductivity([(0,0)-
(2,2)],(0,0),(1,1),(1.27829,1.20175),(1.4,1.2),(1.6,0.947368),(1.78593,0.710526),(2,0.5))
Units: Dmnl
f Effect of WorkLoad Ratio on Quality(((0,0)-
(2,2)],(0,1.2),(0.2,1.2),(0.4,1.17),(0.6,1.12),(0.8,1.06),(1,1),(1.2,0.8736),(1.4,0.712),(1.6,0.489),(1.8,0.3),(2,
0.15))
Units: Dmnl
f pressure to increase training staff([(0,0)-
(2,0.1)],(0,0.02),(0.25688 1 ,0.0157895),(0.391437,0.0122807),(0.556575,0.00701754),(0.715596,0).(1,0),(2
20)
Units: Dmnl
f Pressure to Reduce Staff({(0,0)-(2,1)],(0,0.2),(1,0.75),(2,1))
Units: Dmnl
f Pressure to reduce training staff([{(0,0)-(2,1)],(0,0),(1,0),(2,1))
Units: Dmnl
Fraction Experienced Staff Desired in Training=0.2
26
I. Martinez, G. Wadhwa, and R. MacDonald
Units: Dmnl
(048) Fraction of Experienced Staff in Training="Experienced Staff ( Training)"/Total Experienced Staff
Units: Dmnl
(049) "Fraction of referring base loss/month"=0.1
Units: Dmnl/Month
(050) Fraction patient High=Patients High/Total Patients
Units: Dmnl
(051) Fraction patient low=Patients Low/Total Patients
Units: Dmnl
(052) Fraction patient Medium=Patients Medium/Total Patients
Units: Dmnl
(053) | Gaining experience=Inexperienced Staff/"Average time to get experience/staff"
Units: Staff/Month
(054) Gaining New referrals=Effect of Reputation on new referrals*Current Contact Per Month
Units: Referring people/Month
(055) In patient High=Current Fraction of High patients*New referred patients
Units: Patients/Month
(056) In Patients low=Current Fraction of Low patients*New referred patients
Units: Patients/Month
(057) In Patients Medium=Current Fraction of Medium patients*New referred patients
Units: Patients/Month
(058) Incoming task=Inflow rate*" Average Tasks/patient"
Units: Tasks/Month
(059) Increase in training benefits=Staff Learning
Units: Learning/Month
(060) Increase Staff Providing Training=Staff moved to training* pressure to increase training staff
Units: Staff/Month
(061) — Inexperienced Staff= INTEG ((+New hiring rate-Gaining experience),0)
Units: Staff
(062) Inflow rate=Total Patients in
Units: Patients/Month
(063) Length of Employment=50
Units: Month
(064) Level Of Training= INTEG (Increase in training benefits-Training benefits lost when staff leaves,1677)
Units: Learning
(065) Loss of referral=Effect of Reputation on Loss of referral*"Fraction of referring base loss/month"
*Referring Base
Units: Referring people/Month
(066) Minimum Required Quarterly Profits=1000
Units: Dollars/Quarter
(067) Months Per Quarter=3
Units: Months/Quarter
(068) Net Change in Quarterly Profits=(Estimated Quarterly Profits-Average Quarterly Profits)/Adjustment Time
for Quarterly Profits
Units: Dollars/Quarter/Month
(069) Net Income=Total Revenue-Total staff Cost-Other Cost
Units: Dollars/Month
(070) New hiring rate=Quitting*Pressure to Reduce Staff
Units: Staff/Month
(071) New referred patients=Referring Base*" Average patients referred/month/referring people"
Units: Patients/Month
(072) "Normal Contacts with People/Month"=22
Units: Referring people/Month
(073) | Normal WorkLoad=118
Units: Tasks/Staff
(074) Other Cost=50000
27
(075)
(076)
(077)
(078)
(079)
(080)
(081)
(082)
(083)
(084)
(085)
(086)
(087)
(088)
(089)
(090)
(091)
(092)
(093)
(094)
(095)
(096)
(097)
(098)
(099)
I. Martinez, G. Wadhwa, and R. MacDonald
Units: Dollars/Month
Out Patient High=Average Out patient per group*Fraction patient High
Units: Patients/Month
Out Patient low=Average Out patient per group*Fraction patient low
Units: Patients/Month
Out patient Medium=Average Out patient per group*Fraction patient Medium
Units: Patients/Month
Outflow rate=Completion rate*"Avg. Backlog to tasks"
Units: Patients/Month
Patients High= INTEG (In patient High-Out Patient High,110)
Units: Patients
Patients Low= INTEG (In Patients low-Out Patient low,110)
Units: Patients
Patients Medium= INTEG (In Patients Medium-Out patient Medium, 110)
Units: Patients
Perception of Quality=SMOOTH(Current Quality of Services, Adjustment Time for Perception of Quality)
Units: Quality
pressure to increase training staff=f pressure to increase training staff(Ratio of Avg to Desired level of
training)
Units: Dmnl
Pressure to Reduce Staff=f Pressure to Reduce Staff(Ratio of Average Quarterly Profits to Min Reqd)
Units: Dmnl
Pressure to reduce training staff=f Pressure to reduce training staff(Ratio of Avg to Desired level of
training)
Units: Dmnl
Productivity=Productivity Normal*Effect of WorkLoad onProductivity
Units: (Tasks/Staff)/Month
Productivity Normal=118
Units: (Tasks/Staff)/Month
Quality of Services normal=1
Units: Quality
Quitting=Experienced Staff/(Length of Employment*Pressure to Reduce Staff)
Units: Staff/Month
Ratio of Average Quarterly Profits to Min Reqd=Average Quarterly Profits/Minimum Required Quarterly
Profits
Units: Dmnl
Ratio of Avg to Desired level of training=Average Training per employee/Desired training level
Units: Dmnl
Ratio of Perceived Quality to Normal Quality=Perception of Quality/Quality of Services normal
Units: Dmnl
Reduction in staff training="Experienced Staff ( Training)"*Pressure to reduce training staff/Adjustment
time for moving staff
Units: Staff/Month
Referring Base= INTEG (+Gaining New referrals-Loss of referral,110)
Units: Referring people
Reputation=Ratio of Perceived Quality to Normal Quality
Units: Dmnl
Staff Learning=("Experienced Staff ( Training)"*Trainer training Productivity)/Time for Learning to sink
in
Units: Learning/Month
Staff moved to training=Experienced Staff*(Fraction Experienced Staff Desired in Training-Fraction of
Experienced Staff in Training)/Adjustment time for moving staff
Units: Staff/Month
Staff moving to training functions=-Reduction in staff training Increase Staff Providing Training
Units: Staff/Month
Staff to hire=Desired Staff-Total Staff
28
(100)
(101)
(102)
(103)
(104)
(105)
(106)
(107)
(108)
(109)
(110)
(111)
(112)
(113)
(114)
(115)
(116)
(117)
(118)
(119)
I. Martinez, G. Wadhwa, and R. MacDonald
Units: Staff
Step in tasks=-0.2
Units: Dmnl
Tasks= INTEG (+Incoming task-Completion rate,3300)
Units: Tasks
Time for Learning to sink in=2
Units: Month
Time to hire=1
Units: Month
Time to step in tasks=3
Units: Month
Total Experienced Staff="Experienced Staff ( Training)"+Experienced Staff
Units: Staff
Total Fraction=Desired Fraction of High patients+Desired Fraction of Low patients+Desired Fraction of
Medium patients
Units: Dmnl
Total Patients=Patients High+Patients Low+Patients Medium
Units: Patients
Total Patients in=In patient HightIn Patients low+In Patients Medium
Units: Patients/Month
Total Productive Staff=Experienced Staff+Inexperienced Staff
Units: Staff
Total Revenue="Total Revenue/Month from High"+"Total Revenue/month from Low"+"Total
Revenue/Month from Medium"
Units: Dollars/Month
"Total Revenue/Month from High"="$/Patient High"*Out Patient High
Units: Dollars/Month
"Total Revenue/month from Low"="$/patient Low"*Out Patient low
Units: Dollars/Month
"Total Revenue/Month from Medium"="$/Patient Medium"*Out patient Medium
Units: Dollars/Month
Total Staff="Experienced Staff ( Training)"+Total Productive Staff
Units: Staff
Total staff Cost=Total Staff*"Avg. Salary per staff per month"
Units: Dollars/Month
Trainer training Productivity=10
Units: Learning/Staff
Training benefits lost when staff leaves=0
Units: Learning/Month
WorkLoad=Tasks/Total Productive Staff
Units: Tasks/Staff
WorkLoad Ratio=WorkLoad/Normal WorkLoad
Units: Dmnl
TSS Sd odoedndncbcideiiickickiccicck
-Control
Jnrddbeeorobbrciobeickiib ibe
Simulation Control Parameters
(121)
(122)
(123)
(124)
FINAL TIME = 24
Units: Month
INITIAL TIME =0
Units: Month
SAVEPER = TIME STEP
Units: Month [0,?]
TIME STEP = 0.0625
Units: Month [0,?]
29
I. Martinez, G. Wadhwa, and R. MacDonald
(2) Operations 1 Model
-Control
Jnbdddecceobbcobotikiibca dobre
(02)
(03)
(04)
(05)
Simulation Control Parameters
FINAL TIME = 100
Units: Month
INITIAL TIME =0
Units: Month
SAVEPER = 1
Units: Month [0,?]
TIME STEP = 0.0625
Units: Month [0,?]
Je Sd deca idei dice icbck
.Gary's new
Jncddeeoaobbc abet dbo
(07)
(08)
(09)
(10)
(ayy
(12)
(13)
(14)
(15)
(16)
ay
(18)
(19)
(20)
21)
(22)
Additional Rework Required=Total Implant Completions*(1-Initial Fraction Implant Patients Satisfied)
Units: Patients/Month
Additional Rework Tasks=Additional Rework Required*Tasks Required Per Rework Patient
Units: Tasks/Month
Additions to Backlog of Traditional Patient Tasks=New Patients*Initial Tasks Per Traditional Patients
Units: Tasks/Month
Average Patients Per Task=Backlog of New Procedure Patients/Tasks Per New Procedure Patient
Units: Patients/Tasks
Average Patients Per Traditional Tasks="Backlog of Patients (Traditional Services)"/Backlog Of
Traditional Patient Tasks
Units: Patients/Tasks
Average Tasks Per Patient=Backlog Of Traditional Patient Tasks/"Backlog of Patients (Traditional
Services)"
Units: Tasks/Patients
Average Time to Complete Implant Work=2
Units: Month
Backlog of New Procedure Patients= INTEG (+New Procedures Patients Entering Practice-New Procedure
Patients Receiving Services,40)
Units: Patients
"Backlog of Patients (Traditional Services)"= INTEG (New Patients-Patients Leaving-"Patients Leaving -
Services Completed",200)
Units: Patients
Backlog Of Traditional Patient Tasks= INTEG (Additions to Backlog of Traditional Patient Tasks-
Traditional Tasks Completed-Task Reduction Due to People Leaving Before Procedures, Initial Tasks Per
Traditional Patients*"Backlog of Patients (Traditional Services)")
Units: Tasks
Delay in Receiving Services="Backlog of Patients (Traditional Services)"/"Patients Leaving - Services
Completed"
Units: Month
Desired Combined Backlog=1 .238
Units: Month
Desired Hours Per Month Worked by Gary=174
Units: Hours/Month
Desired Service Delivery Delay for Implant Patients=2
Units: Month
Desired Waiting Time for Services=2
Units: Month
Effect of Delay Ratio on Patients Leaving=f Effect of Delay Ratio on Patients Leaving (Ratio of Actual to
Desired Waiting Time)
30
(23)
(24)
(25)
(26)
27)
(28)
(29)
(30)
GB)
(32)
(33)
(G34)
(35)
(36)
G7)
(38)
G39)
(40)
(41)
(42)
(43)
I. Martinez, G. Wadhwa, and R. MacDonald
Units: Dimensionless/Month
Effect of Implant Service Delivery Delay On Patients Willingness to Wait
=f Effect of Implant Service Delivery Delay On Patients Willingness to Wait
(Ratio of Actual Delay to Desired Delay for Implant Patients)
Units: Dimensionless
f Effect of Delay Ratio on Patients Leaving([(0,0)-
(4,1)],(0,0),(1,0),(1.26829,0.0344828),(1.61672,0.132184),(2.02091
,0.327586),(2.31359,0.471264),(2.73171,0.568965),(2.99652,0.603448),(4,0.6))
Units: Dimensionless/Month
f Effect of Implant Service Delivery Delay On Patients Willingness to Wait
({(0,0)-(5,1)],(0,1),(1,1),(1.50523,0.954023),(1.89547,0.867816),(2.34146,
0.568965),(2.63415,0.373563),(2.91289,0.212644),(3.14983,0.103448),(3.41463
,0.0172414),(4,0),(5,0))
Units: Dimensionless
Fraction of Procedure 1 Performed=Initial Fraction of Procedure 1 Performed/Total Fraction of Procedures
Units: Dimensionless
Fraction of Procedure 2 Performed=Initial Fraction of Procedure 2 Performed/Total Fraction of Procedures
Units: Dimensionless
Fraction of Procedure 3 Performed=Initial Fraction of Procedure 3 Performed/Total Fraction of Procedures
Units: Dimensionless
Fraction of Procedure 4 Performed=Initial Fraction of Procedure 4 Performed/Total Fraction of Procedures
Units: Dimensionless
Fraction of Procedure 5 Performed=Initial Fraction of Procedure 5 Performed/Total Fraction of Procedures
Units: Dimensionless
Fraction of Time Spent on Implant Patients=Hours Spent With Implant Patients/Total Hours Allocated Per
Month
Units: Dimensionless
Fraction of Time Spent on Management Issues=Hours Spent on Management Issues/Total Hours Allocated
Per Month
Units: Dimensionless
Fraction of Time Spent on Traditional Patients=Hours Spent on Traditional Patients/Total Hours Allocated
Per Month
Units: Dimensionless
"Fraction of Time Spent Training Others (Staff and Other Doctors)"="Hours Spent Training Others (Staff
and Other Doctors)"/Total Hours Allocated Per Month
Units: Dimensionless
"Fraction of Time Spent With Implant Patients (Option Two)"=Hours Spent With Implant Patients/Total
Patient Contact Hours
Units: Dimensionless
"Fraction of Time Spent With Traditional Patients (Option Two)"=Hours Spent on Traditional
Patients/Total Patient Contact Hours
Units: Dimensionless
Hours Spent on Management Issues=40
Units: Hours/Month
Hours Spent on Traditional Patients=108
Units: Hours/Month
"Hours Spent Training Others (Staff and Other Doctors)"=2
Units: Hours/Month
Hours Spent With Implant Patients=24
Units: Hours/Month
Implant Patients Waiting for Completion by Their Dentist= INTEG (+New Procedure Patients Receiving
Services-Additional Rework Required-Satisfactory Implant Completions,40)
Units: Patients
Initial Fraction Implant Patients Satisfied=1
Units: Dimensionless
Initial Fraction of Procedure | Performed=0.05
31
44)
45)
(46)
7)
48)
9)
(50)
G1)
(52)
(53)
(54)
(55)
(56)
(67)
(58)
(659)
(60)
(61)
(62)
(63)
(64)
(65)
(66)
(67)
I. Martinez, G. Wadhwa, and R. MacDonald
Units: Dimensionless
Initial Fraction of Procedure 2 Performed=0.7
Units: Dimensionless
Initial Fraction of Procedure 3 Performed=0.1
Units: Dimensionless
Initial Fraction of Procedure 4 Performed=0.05
Units: Dimensionless
Initial Fraction of Procedure 5 Performed=0.1
Units: Dimensionless
Initial Tasks Per Traditional Patients=10
Units: Tasks/Patients
Losses from the Referral Base=0
Units: Doctors/Month
Months of Combined Patient Backlog=Total Backlog of Tasks/"Total Tasks Capable of Being Performed
Per Month - All Patients"
Units: Month
New Additions to Referral Base=0
Units: Doctors/Month
New Patients=Traditional Patients Referred to Practice
Units: Patients/Month
New Patients Referred to Practice=Referral Base*Patients Referred by Referral Base Doctors Per Month
Units: Patients/Month
New Procedure Patients=Fraction of Procedure 5 Performed*New Patients Referred to Practice
Units: Patients/Month
New Procedure Patients Receiving Services=New Procedure Tasks Completed* Average Patients Per Task
Units: Patients/Month
New Procedure Tasks Completed=Tasks Performed Per Month on Implant Patients
Units: Tasks/Month
New Procedures Patients Entering Practice=New Procedure Patients
Units: Patients/Month
New Tasks Associated with New Procedure Patients=Tasks Per New Patient Procedures*New Procedures
Patients Entering Practice
+Additional Rework Tasks
Units: Tasks/Month
Patients Leaving="Backlog of Patients (Traditional Services)"*Effect of Delay Ratio on Patients Leaving
Units: Patients/Month
"Patients Leaving - Services Completed"=Traditional Tasks Completed* Average Patients Per Traditional
Tasks
Units: Patients/Month
Patients Referred by Referral Base Doctors Per Month=2
Units: Patients/Doctors/Month
Ratio of Actual Delay to Desired Delay for Implant Patients=Service Delivery Delay for New Procedure
Patients/Desired Service Delivery Delay for Implant Patients
Units: Dimensionless
Ratio of Actual Hours Per Month to Desired Hours Per Month=Total Hours Allocated Per Month/Desired
Hours Per Month Worked by Gary
Units: Dimensionless
Ratio of Actual To Desired Backlog=Months of Combined Patient Backlog/Desired Combined Backlog
Units: Dimensionless
Ratio of Actual to Desired Waiting Time=Delay in Receiving Services/Desired Waiting Time for Services
Units: Dimensionless
Referral Base= INTEG (+New Additions to Referral Base-Losses from the Referral Base, 100)
Units: Doctors
Satisfactory Implant Completions=Total Implant Completions* Initial Fraction Implant Patients Satisfied
Units: Patients/Month
32
I. Martinez, G. Wadhwa, and R. MacDonald
(68) Service Delivery Delay for New Procedure Patients=Backlog of New Procedure Patients/New Procedure
Patients Receiving Services
Units: Month
(69) Task Reduction Due to People Leaving Before Procedures=Patients Leaving*Average Tasks Per Patient
Units: Tasks/Month
(70) Tasks Per New Patient Procedures=15
Units: Tasks/Patients
(71) Tasks Per New Procedure Patient= INTEG (+New Tasks Associated with New Procedure Patients-New
Procedure Tasks Completed,Backlog of New Procedure Patients*Tasks Per New Patient Procedures)
Units: Tasks
(72) "Tasks Per Patient Hour (Implant Patient)"=12.5
Units: Tasks/Hour
(73) "Tasks Per Patient Hours (Traditional Patients)"=16.6667
Units: Tasks/Hour
(74) Tasks Performed Per Month on Implant Patients="Tasks Per Patient Hour (Implant Patient)"*Hours Spent
With Implant Patients
Units: Tasks/Month
(75) Tasks Performed Per Month on Traditional Patients="Tasks Per Patient Hours (Traditional
Patients)"*Hours Spent on Traditional Patients
Units: Tasks/Month
(76) Tasks Required Per Rework Patient=5
Units: Tasks/Patients
(77) Total Backlog of Patients=Backlog of New Procedure Patients+"Backlog of Patients (Traditional
Services)"
Units: Patients
(78) Total Backlog of Tasks=Tasks Per New Procedure Patient+Backlog Of Traditional Patient Tasks
Units: Tasks
(79) Total Fraction of Procedures=Initial Fraction of Procedure 1 Performed+Initial Fraction of Procedure 2
Performed+Initial Fraction of Procedure 3 Performed+Initial Fraction of Procedure 4 Performed+Initial
Fraction of Procedure 5 Performed
Units: Dimensionless
(80) Total Hours Allocated Per Month=Hours Spent on Management Issuest+Hours Spent on Traditional
Patients+"Hours Spent Training Others (Staff and Other Doctors)"
+Hours Spent With Implant Patients
Units: Hours/Month
(81) Total Implant Completions=Implant Patients Waiting for Completion by Their Dentist/Average Time to
Complete Implant Work
Units: Patients/Month
(82) Total Patient Contact Hours=Hours Spent on Traditional Patients+Hours Spent With Implant Patients
Units: Hours/Month
(83) "Total Tasks Capable of Being Performed Per Month - All Patients"=Tasks Performed Per Month on
Traditional Patients+Tasks Performed Per Month on Implant Patients
Units: Tasks/Month
(84) Traditional Patients Referred to Practice=New Patients Referred to Practice*(Fraction of Procedure 1
Performed+Fraction of Procedure 2 Performed
+Fraction of Procedure 3 Performed+Fraction of Procedure 4 Performed)
Units: Patients/Month
(85) Traditional Tasks Completed=Tasks Performed Per Month on Traditional Patients
Units: Tasks/Month
(3) Financial 1 Model
-Control
Jee ediiduidcidcindcabcdeidaiciccicccck
Simulation Control Parameters
33
I. Martinez, G. Wadhwa, and R. MacDonald
(02) FINAL TIME = 1000
Units: Day
(03) INITIAL TIME =0
Units: Day
(04) | SAVEPER =1
Units: Day [0,?]
(05) TIME STEP = 0.03125
Units: Day [0,?]
JncdeSorobnerobeciiebcei dobre
Gary's ar
Jecddecorobbciobcciiiebicks dobre
(07) "30 Days"=30
Units: Day
(08) "AIR Aging 1"="Outflow from A/R 30"*(1-"Fraction Paid A/R 30")
Units: Units/Day
(09) "A/R Aging 2"="Outflow from A/R 60"*(1-"Fraction Paid A/R 60")
Units: Units/Day
(10) "A/R Aging 3"="Outflow for A/R 90"*(1-"Fraction Paid A/R 90")
Units: Units/Day
(11) "AIR Aging 4"="Outflow for A/R 120"*(1-"Fraction Paid A/R 120")
Units: Units/Day
(12) "A/R Greater than 120"= INTEG (+New Bills Not Paid in 120 Days-"A/R Written off",New Bills Not Paid
in 120 Days*"Time to Write off A/R")
Units: Units
(13) "AIR Paid for 0 to 30 Days"="Outflow from A/R 30"*"Fraction Paid A/R 30"
Units: Units/Day
(14) "AIR Paid for 120 Day"="Outflow for A/R 120"*"Fraction Paid A/R 120"
Units: Units/Day
(15) "AR Paid for 31 to 60 Days"="Outflow from A/R 60"*"Fraction Paid A/R 60"
Units: Units/Day
(16) "A/R Written off"="A/R Greater than 120"/"Time to Write off A/R"
Units: Units/Day
(17) "Accounts Receivable (0 to 30)"= INTEG (+New Bills-"A/R Aging 1"-"A/R Paid for 0 to 30 Days",New
Bills*"30 Days")
Units: Units
(18) "Accounts Receivable (31 to 60)"= INTEG ("A/R Aging 1"-"A/R Aging 2"-"A/R Paid for 31 to 60
Days","A/R Aging 1"*"30 Days")
Units: Units
(19) "Accounts Receivable (61 to 90)"= INTEG ("A/R Aging 2"-"A/R Aging 3"-AR Paid for 90 Day,"A/R
Aging 2"*"30 Days")
Units: Units
(20) "Accounts Receivable (90 to 120)"= INTEG ("A/R Aging 3"-"A/R Aging 4"-"A/R Paid for 120 Day","A/R
Aging 3"*"30 Days")
Units: Units
(21) AR Paid for 90 Day="Outflow for A/R 90"*"Fraction Paid A/R 90"
Units: Units/Day
(22) "Fraction Paid A/R 120"=0.1
Units: Dimensionless
(23) "Fraction Paid A/R 30"=0.8+step(0.15,25)
Units: Dimensionless
(24) "Fraction Paid A/R 60"=0.7
Units: Dimensionless
(25) "Fraction Paid A/R 90"=0.6+step(0.35,100)
Units: Dimensionless
(26) New Bills=30
Units: Units/Day
34
27)
(28)
(29)
(30)
GB)
(32)
(33)
I. Martinez, G. Wadhwa, and R. MacDonald
New Bills Not Paid in 120 Days="A/R Aging 4"
Units: Units/Day
"Outflow for A/R 120"="Accounts Receivable (90 to 120)"/"30 Days"
Units: Units/Day
"Outflow for A/R 90"="Accounts Receivable (61 to 90)"/"30 Days”
Units: Units/Day
"Outflow from A/R 30"="Accounts Receivable (0 to 30)"/"30 Days"
Units: Units/Day
"Outflow from A/R 60"="Accounts Receivable (31 to 60)"/"30 Days"
Units: Units/Day
"Time to Write off A/R"=245
Units: Day
Total Late Accounts Receivable="Accounts Receivable (90 to 120)"+"Accounts Receivable (61 to
90)"+"Accounts Receivable (31 to 60)"
+"A/R Greater than 120"
Units: Units
35
Back to the
