Improving Professional Practice for Mobile Operators: Decision Support Tool for Outsourcing

10180 words (41 pages) Dissertation

9th Dec 2019 Dissertation Reference this

Tags: Outsourcing

Disclaimer: This work has been submitted by a student. This is not an example of the work produced by our Dissertation Writing Service. You can view samples of our professional work here.

Any opinions, findings, conclusions or recommendations expressed in this material are those of the authors and do not necessarily reflect the views of

  The pilot study of the current research was the first step of the practical application of a decision support tool for outsourcing. In this essay the researcher covers a theoretical background of the definition and value of a pilot studies. He also covers the goal of a pilot study – what he expects from a pilot study. The researcher then gives an overview of the Theoretical Framework underpinning the research to give context and reference to the reader. The researcher then discusses the application of the pilot study in the current research. Finally, the outcomes of the pilot study will be examined, as they will have a direct influence on the actual research itself. We start with the definition a pilot study and state the value thereof following the introduction to clarify what a pilot study really is and why it is needed in the research process.
  A pilot study is a mini-version of a full-scale study or a trial run done in preparation of the complete study. The latter is also called a ‘feasibility’ study. It can also be a specific pre-testing of research instruments, including questionnaires or interview schedules. (Compare Polit, et al. & Baker in Nursing Standard, 2002:33-44; VanTeijlingen & Hundley, 2001:1.) The pilot study will thus follow after the researcher has a clear vision of the research topic and questions, the techniques and methods, which will be applied, and what the research schedule will look like. It is “reassessment without tears” (Blaxter, Hughes & Tight, 1996:121), trying out all research techniques and methods, which the researcher have in mind to see how well they will work in practice. If necessary, it can then still be adapted and modified accordingly. (Blaxter, Hughes & Tight, 1996:121) The pilot study in this current research can be defined as mainly a try-out of research techniques and methods. The researcher created a system dynamic model and applied this to a pilot group of employees and network KPIs. The value of first piloting the whole research process is discussed in the next section, because if a pilot study is of too little value, the researcher can waste time, energy and money.
The researcher will discuss first the value of a pilot study as explained by different authors and then the applicability to the current study in the following paragraphs. After stating the value of such a study, the researcher will compile the goal of a pilot study for the current research project.
  1. The Value of a Pilot Study 
Things never work quite the way you envisage, even if you have done them many times before, and they have a bad habit of turning out very differently than you expected. “You may think that you know well enough what you are doing, but the value of pilot research cannot be overestimated” Blaxter, et al. (1996:122) It is thus obvious to the researcher, that the pilot study in the current research was essential to prevent the waste of time, energy and money. The value is also emphasised by the points listed below. Pilot studies could be conducted in qualitative, quantitative, and even mixed methods research. General application of pilot studies can be summarized in four areas: 1) To find problems and barriers related to participants' recruitment. 2) Being engaged in research as a qualitative researcher. 3) Assessing the acceptability of observation or interview protocol. 4) Determine epistemology and methodology of research. A pilot study can explore the limitations of access to data due to cultural sensitivities. Also It can help researchers with refining the sampling strategy. In fact, the pilot study in addition to providing a ground for self-assessment of researchers' preparation and capacity could help them to practice qualitative inquiry and as a consequence enhance the credibility of a qualitative research. Considering that the development of a plan for data collection requires researcher's insight and creativity beyond a mechanical inquiry in frame of the research questions and model development, conducting a pilot study could facilitate judgment about the possibility of obtaining sufficient and rich data to answer the research question as well. The pilot study in the current research process was very specifically used to identify practical problems in the process, sessions and methods used. The research itself has as a goal the applicability of the developed decision support tool for outsourcing in the Telecommunication mobile industry to improve their professional practice The pilot study would thus indicate whether the proposed methods and / or instruments are appropriate. The pilot study could also give advance warning of possibilities where certain types of techniques or the study as a whole could fail. A pilot study can therefore be of value for testing the feasibility of either research instruments or data collection instruments and also of the research process itself. The following section combined the statements of the value of pilot studies in a goal of pilot studies in general as well as for the current research project.
  1. The Goal of a Pilot Study
  The researcher sees the goal of a pilot study in general as related to the aim of the research project of which it forms part. The general goal of a pilot study is to provide information, which can contribute to the success of the research project as a whole. The goal is thus to test the study on small scale first to sort out all the possible problems that might lead to failure of the research procedure. It might minimise the risk of failure. In the current study the goal of the pilot study consists of two parts:
  1. To find as many as possible practical arrangements that might have a negative influence on the success of the research procedure.
  2. The other included sorting out all practicalities related to measurement instruments as well as the applicability of these instruments to the potential outcomes of the study.
The procedure of the pilot study in the current research project is discussed in the following paragraphs.
The pilot study of the current research follows the design phase, which is the research strategy as stated in Chapter 1. Since the participants in outsourcing decision making incompletely assess the use of resources (work, capital and knowledge) and capabilities in the framework model of company policy for the establishment of relationship with the outsourcer, I tried to find out why enterprises decide to enter outsourcing relationships. I had a number of questions that I wanted to explore:
  • Why many outsourcing agreements failed to achieve part or whole of its goals?
  • How to avoid the significant risk to the company when it’s outsourcing strategy failed?
  • How to facilitate such an important decision that many operators are facing frequently which could change their entire strategy and positioning?
Yes, the failure of outsourcing agreement impact is not only limited to not achieving the target cost savings, efficiency and better performance, but it goes beyond that up to getting the company out of competition! From here it came up to the researcher the idea of developing a “Decision support tool for outsourcing” Three basic premises given below have arisen while thinking about the purpose of the research: 1.The choice of partners – higher complexity, specialisation and the division of labour make it possible for the outsourcers to carry out several activities with lower costs and a higher added value, than in the case of carrying out all activities inside the company; the outsourcing company chooses suppliers, who improves the outsourcers position on the market through their knowledge, capabilities and technology. 2. The consequences of short-term placement – the majority of companies are in favour of short-term and mainly financial results of the outsourcing relationship. They are rarely aware of the long-term consequences of their actions, thus it is necessary to study the use of capabilities brought about by establishing and termination outsourcing activities and compare them in the temporal framework with the benefits for both the outsourcer and outsources. 3. The consequences of eventual termination of outsourcing – the review of literature shows that enterprises rarely deal with problems, which may arise if a company decides to terminate outsourcing activity and brings it back to the outsourcing company. Thus, it is necessary to find out if the outsourcing company still has the equipment and professional staff, who is familiar with the process, financial assets etc. or the position in which the outsourcee may find itself in. Furthermore, we should analyse difficulties of the two companies, the outsourcer and outsourcee at the time of termination of outsourcing activity.
  - Secondary data collection - Focus group discussions
  In recent years competition has been fierce between the three mobile operators in Egypt (Vodafone Group-backed Vodafone Egypt, Orange-backed Mobinil and Etisalat-backed Etisalat Misr). This has been primarily due to the increased market saturation and the implementation of new regulatory policies such as mobile number portability. New market developments, such as the proposed introduction of Mobile Virtual Network Operators (MVNO) services & introduction of fourth operator have intensified the competition, inevitability putting pressure on the operators’ costs and profitability, which have already been impacted by the recent political events in Egypt including and not limited to 2011 and 2013 revolutions. As a result of changing market dynamics: reducing costs and maintaining profitability are the key issues on top of Egyptian mobile operators’ agendas. Outsourcing has emerged as an effective strategic tool for these operators to address the key issues of cost reduction and improved profitability. The unrelenting pressure for greater efficiencies has forced many firms to increasingly focus on their core competencies and hence, specialize in a limited number of key areas. This has led operators to outsource some activities that traditionally have been carried out in-house. A number of factors make the intersection of mobile telecommunications industry, the MENA geography (represented by Egypt) and Outsourcing a prime research topic, among them: -The mobile telecommunication industry is booming in Middle East and North Africa (MENA) where Egypt represents the largest market in the Arab world. -Egypt’s mobile market is among the most competitive in MENA and is key driver for its commerce. -Many of the studies of outsourcing in the telecommunications industry have focused primarily on the motives for outsourcing and have failed to provide an in-depth understanding on the outcomes associated with outsourcing.

6.1  Overview of  the Mobile Telecommunications Business in Egypt

Egypt's mobile subscriber base had risen to 105.5 million from 83.121 million at the end of December 2015 (Table 1). The telecommunication industry of Egypt is one of the fastest developing sectors in the country, by 2017; it is believed that the market will have around 110.9mn subscriptions and penetration rate of 125.2%. Since the establishment of the Ministry of Communications and Information Technology in 1999, the Egyptian telecom industry has been ushered into an era of liberalized policies and new regulatory laws. In recent times, there has been a significant growth in low-income customer segment owing to stiff competition between various mobile operators, which has led to tariff reductions. Egypt's mobile market is among the most competitive in the Middle East and North Africa region, playing host to four major international players: The Vodafone Group-backed Vodafone Egypt, Orange-backed Orange-Egypt and Etisalat-backed Etisalat Misr plus the new awarded  WE backed up by Telecom Egypt - see Table (2). Business environment became more competitive, especially with market saturation, the introduction of fourth operator, the Mobile-Fixed Convergence and the implementation of new regulatory policies such as mobile number portability. We expect competition to become even more intense in the next few years following the proposed introduction of MVNO, FVNO and 4G services. Table (1): Mobile Market Overview Clearly, the political situation in Egypt has had a positive effect on mobile subscriber growth contrary to previous expectations. However, the effect on operators' financial indicators is less encouraging as Orange Egypt  and Vodafone Egypt reported a sharp decline in ARPUs and, consequently, net profits during past years (figure 1). Table (2): Competitive Landscape Source: BMI Egypt Infrastructure Report Q4 2016, Publish By: Business Monitor international 2016     There are still a number of negative characteristics of Egypt's telecoms sector. These include a mobile market that is highly skewed towards prepaid users and falling mobile ARPUs. In addition, price competition has been aggressive since the introduction of compulsory SIM registration in May 2010. Although Egypt's operators have reported sequential increases in ARPU, the overall trend in the market remains downwards as strong competition continues to give consumers large amounts of choice and forces prices down. Figure (1): Industry Forecast – ARPU 2011-2018, BMI Forecast, Source BMI

6.2  Motivation

A number of questions motivated the researcher to start his research: How can managers of Egyptian Telecommunication operators successfully deal with fierce competition? Can Outsourcing be an effective strategic tool for these operators to address cost reductions and improved profitability through right sourcing decision? What are the key factors that affect a sourcing strategy decision? What is the appropriate sourcing strategy that to be pursued by business leaders in the Egyptian telecommunication sector? And Why?

6.3  Problem Statement

With a population of 90+ millions, Egypt is the largest market in the Arab world. However, unemployment by 2018 is high at 15% a phenomenon that subdues demand. Furthermore, Egyptian wages are low in global terms, and, though this offers certain advantages to foreign investors, it also implies that there are limited opportunities to rapidly deploy more lucrative, high margin, telecoms products. Meanwhile, Egypt has a relatively low level of urbanization, with only around 43% of the population living in towns and cities. This presents telecoms network operators with specific challenges when it comes to rapidly extending new services and technologies to the wider populace. Furthermore, Egypt's mobile operators are starting to show concern over the continued downtrend of the Average Revenue Per User (ARPU) and their weak subscriber mixes. To further compound the challenge facing the operators in Egypt, in addition to competing on the price of telecom services, they are also expected to compete in other key areas such network quality and completeness of coverage. This view is supported by the fact that, in spite of the operators’ margin compression, there were several large-scale network upgrades and expansion plans announced by the mobile operators during the last 5 years starting 2013. In the face of these challenges, Telecom operators are studying the adoption of an outsourcing strategy to reduce some of the operational burdens. However, their profitability in the longer term also depends on their ability to innovate, in addition to reducing their operational costs. The problem that the research is tackling is to identify relation between the adoption of outsourcing strategy in telecom operators and their profitability through delivering competitive network quality with optimized cost. Hence the research main question will be: How to improve the professional practice of mobile operators in the sourcing domain?

6.4  Research Objectives

When starting to think about this research , from professional practices many outsourcing decisions have failed leaving the operator bleeding and not able to provide adequate service. Even the operator is not capable of bringing back his skilful resources that was sacrificed after an outsourcing decision that was not really well analysed! Hence, the ultimate objective of this research was How to facilitate such an important decision of outsourcing that many operators are facing frequently which could change their entire strategy and positioning? Is it possible to provide a simple tool that help the executives and practitioners in the mobile operators to help them in making an adequate decision with minimal risk? This research will also help guide telecom operators to solve the long-standing questions regarding adopting a sound outsourcing strategy. This will cover the academic and practitioner viewpoints: Building a model to explore the nature of relationship between all addressed constructs namely: strategic sourcing, mobile telecom operator (market), organization performance (Network quality, costs,…) and innovation. This model will enable the examination of how the choice of the outsourcing strategy affects the organization performance.

6.5  About the Researcher:

The researcher’s is currently a Senior Director of the Network at Orange Egypt, with a cumulative experience of over 20 years in the field of telecommunications. He currently oversees budgeting and dimensioning of the entire Orange network.  His background includes strategic planning for start-up engineering projects, operations, maintenance, human resources balancing, and finance. He successfully grew revenue, increased efficiency and productivity, reduced costs, improved operations, and expanded the company footprint to 30+ Million subscribers. He is a results-driven executive with a solid understanding of the practices, technologies, and service providers within the Telecom & IT industry. The researcher’s current responsibilities include managing an organization of 300+ employees and an annual operating budget in excess of $150+ million USD. This budget must be deployed judiciously and stretched to maximize its impact. In that role, the decision whether to outsource a certain function or keep it in-house is encountered all-too-often. In our industry, we are inundated with information regarding outsourcing and its benefits; however, information regarding its impact on an operators’ profitability is scarce and hard to come by. With an engineering background the researcher has always attempted to develop an analytical framework for decision-making. It is in this context that the researcher has taken a keen interest in the topic and decided to make it the focus of the DBA research thesis.          
  In this section, I will go quickly through the main theories tackled during the literature review and highlight the main ideas and elements extracted to base the foundation of building my system dynamic modelling. It was found through the literature review that the majority of the outsourcing models are built based on 3 main theories:
  1. Transaction Cost Economics (TCE)
  2. Resource Based View (RBV)
  3. Agency Theory.
A number of outsourcing studies in the telecoms industry have employed either Transaction Cost Economics (TCE) as Edoardo Mollona & Alessandro Sposito (2008), Jacobides, M.G. and Winter, S.G.(2005) Jiang, B., Belohlav, J. Young, S. (2007).   Or Resource-Based View (RBV) as Lowson (2002) Coates and McDermott (2002) Vastag, (2000) Aron, R., Singh, J.V.(2005) Ellram et al.(2008); Youngdahl and Ramaswamy (2008)  or Agency Theory as Logan, Mary S (2000) theoretical frameworks to undertake their analysis. The ideas/concepts extracted from the three aforementioned theories will be used jointly, for my proposed model, which is possible to accomplish with our System Dynamics approach. System Dynamics models are computational representations of the causal structure of systems—be they physical, social, or economic—as a set of differential equations using stock and flow variables. The stock and flow variables are arranged in structures called causal loops to eventually form Causal Loop Diagrams or CLDs . One of the most important reasons to use system dynamics it its capability to manage soft variables included in our model like ” Resource Capabilities” variable, as far as soft variables are concerned, numerical data are often unavailable or non-existent. Despite this, such variables are known to be critical to decision making and, therefore should be incorporated into system dynamics models. The stocks and flows constituting each loop in the researcher’s proposed CLD will have its theoretical underpinning tied to one of the three aforementioned theories, for which we present a short overview next.
  1. The knowledge based view (RBV) theory:
  The Resource Based View (RBV) theory posits that firms create sustained competitive advantage with resources that are rare, valuable, imperfectly imitable and not substitutable (Barney 1991). Firms as constructs of human interaction tend to develop their own language for codifying knowledge and their own routines to enhance internal processes. If an activity is highly specific to a company, it is embedded in the company’s language and routines. Employees are then familiar with this “common organization communication code” (Monteverde 1995). Thus, activities can be governed more efficiently within the firm. RBV then does not predict how efficient an external purchase can be, it rather points out that the more firm-specific an activity/resource, “the greater use it makes of firm-specific language and routines, and hence the more efficient is internal governance” (Poppo and Zenger 1998, p. 858). From there, It was taken into consideration the importance of the  effect of the knowledge/Experience of the firm’s employees into its performance, and hence the outsourcing decision.
  1. Transaction costs economics (TCE) theory:
  Following transaction costs economics (TCE), external suppliers can achieve production cost efficiencies through economies of scale and specialization (Marshall et al. 2007), which provides a motive for outsourcing (Poppo and Zenger 1998). However, other costs related to the exchange of services within or across firm boundaries, such as search, selection, bargaining, monitoring and enforcement (Madhok 2002), may offset the production cost savings of external suppliers given the higher likelihood of opportunistic behavior of an external supplier compared to an internal unit (Williamson 1991). Frequency, asset specificity, and uncertainty are the key drivers of transaction costs. External suppliers build Global Network Operations Centers to serve the networks of many customers. These few but large operation centers presumably work more efficiently than the sum of all small operation centers managed by and serving single network operators. This is in part due to fixed costs, for example in problem-solving teams. Suppliers serving several networks need only one such team as network breakdowns rarely occur simultaneously in multiple networks. Hence, specialized external suppliers can offer network operation services at lower cost than internal departments at operators (Hecker and Kretschmer 2010). For transaction costs, Crandall et al. (2009) argued for the telecommunications market that negotiating issues such as “prices for maintaining the network, connecting subscriber lines, and replacing network elements as they depreciate” (p. 506) are complex and that the efficiency of market governance is likely to be low. We analyze market-related transaction costs in detail based on the three major drivers for transaction costs. First, consider the frequency of transaction. Network operators communicate on a daily basis with their external services partner about technical issues. However, they do not change their service supplier frequently but sign contracts for three to five years, avoiding high costs due to on-going searching and negotiating. Second, we assess asset specificity of network operation services for a network operator. These assets do not have to be of a physical nature (Klein et al. 1978) and can be interpreted as the knowledge and expertise employees of an external supplier have developed. As mentioned earlier, all major network equipment vendors have the ability to manage not only their own equipment, but also infrastructure initially built by a competitor. Both the interfaces between billing systems or customer management databases and the network infrastructure have been standardized since the launch of 2G mobile. Hence, operators can easily switch suppliers, if they need to. Third, Environmental uncertainty primarily refers to the inability to predict market demand (McNally and Griffin 2004), which is constantly “shifting and evolving” in the Telecommunications industry (Crandall et al.2009). Despite this, mobile network operators cannot adapt their physical network quickly to demand fluctuations. In summary, from a TCE perspective it is important to take into consideration the cost of doing the service .Comparing the internal Cost with the outsourcing costs but taking into consideration the RBV effect. Also, the risk of not being able to get the internal resources back again once we decide to go for outsourcing should be considered. As well as the service performance comparison (in our case the network performance comparison) into consideration, which will be highlighted in the next section through the Agency theory.
  1. Agency Theory:
  Agency Theory has a long tradition in analyzing situations when parties cooperate through the division of labor (Eisenhardt 1989). More precisely, it examines situations where a principal delegates work to an agent. The focal point of analysis is how to align the interests of the agent in an efficient and cost-effective way with those of the principal. If an agent’s performance can be measured adequately, market prices provide the most effective incentives for the agent to act in accordance with the principal’s interests (Poppo and Zenger 1998). If the performance of an agent, however, is difficult to measure, market contracts might be less efficient than internalizing the principal-agent relationship (Barzel 1989, p. 76). Within an organization, principals can suppress opportunistic behavior of an agent by “behavioral monitoring” (Poppo and Zenger 1998, p. 859) and the use of authority instead of incentives. In market transactions, such instruments are not available. We now question if the performance of external network operation services can be measured accurately. If all functions related to operation services are outsourced, the focal variable is the overall quality and reliability of the network, which is crucial for the success of an operator. In practice, network operators include key performance indicators, audits and service benchmarks in their contracts with external suppliers and measure network quality via overall network coverage and the number of breakdowns (Friedrich et al. 2009, p. 14). From the Agency theory, we can understand the importance of the service delivered (here the network Key Performance/Quality Indicators) to be taken into consideration as well as the costs and the skills in the outsourcing strategy decision.        
  1. Integrated Framework
  After highlighting the main variables to be taken into consideration going through the main theories governing that domain, I tried to develop a decision support tool for the executives of the Mobile operators to guide and support them to decide on their strategy in managing their Network. Whether outsource or not? why? when? and at which cost? By addressing the application of the ideas extracted from TCE, RBV and Agency theories in an integrative framework ;we can then understand if it is Beneficial for local telecommunication network and service providers to outsource activities to international managed service providers or no. Specifically, the author applies RBV to address the questions related to the strategic importance of those activities, assessing whether they are core competences of the firms or not. By contrast, TCE will assist in assessing whether economic advantages are actually achievable by outsourcing activities. Finally, Agency theory will be used to assess the quality measure. Epistemologically the qualitative paradigm was chosen in this research for several reasons; first, the research main objective is to “explore” how the outsourcing affects the mobile operators’ profitability in Egypt. The researcher is trying to examine the current change in “outsourcing phenomena” that started to rise in the telecommunication field. As the researcher will be more concerned about the quality and texture of sourcing experience, the researcher in this paradigm is trying to explore the relation between the sourcing model and operators’ profitability. The research involves observation, measuring and testing:
  • Critical observation
  • Model Building
  • Analysing
  • Evaluation
  1. Initial causal loop diagram:
  The initial CLD shown below in ( figure 2) was the basic idea of this research , the aim was to give a decision support for the executives and managements of Telecommunication Mobile operator whether to outsource or not their network activities. The start-up idea was to compare the Profits (Revenues-Costs) of those activities when handled in house versus the Profits when they are outsourced. The researcher realised then that there other elements that could impact the customer satisfaction else than the employees experience/performance and the network. Also the same applies on the revenues , as there are a lot of factors that could impact the revenues else than the customer satisfaction like the sales force , the competition and many others… After many researches and discussion with the supervisor and the executives of the company. And based on the long experience of the researcher in the Mobile Telecommunication industry; it was decided to focus mainly on the Network KPIs impact and the sourcing costs. Fig (2) Initial Model CLD 8.2  Model  Overview: The purpose of this model is to help in the decision making of insourcing or outsourcing the engineering services of a given project / rollout based on the following parameters;
  • Availability of the qualified in-house resources.
  • Insourcing cost.
  • Out-sourcing cost.
  • Best Competitor KPIs.
  • Vendor KPIs.
The main focus now in the upcoming sections is how to determine the availability of the qualified in-house resources and the insourcing cost then later on I can build on this to take into account the effect of the Key performance indicators (KPIs) and the outsourcing cost. 8.2.1 Insourcing scenario prediction/estimation: Training hours & Platforms knowledge In order to simulate the availability of in-house resources and their movements (promotions, churn) inside a given organization, the following assumption is made to be able to quantize the process; The promotion of an employee from a level to another depends on the employees’ knowledge of a specific number of platforms and the availability of open positions in that level. The platform could be the products of a specific OEM or Vendor. To be able to measure the knowledge of a given employee, the number of training hours received is considered in this study as the reference for the knowledge of a given number of platforms. At given thresholds of training hours: Training Hours_ Threshold-1 & Training Hours_ Threshold-2, which are user inputs, the knowledge of one, two or three platforms is determined. If the training hours obtained by a given employee are more than Training Hours_ Threshold-1 and less than Training Hours_ Threshold-2 then this employee knows 2 platforms and if exceeded Training Hours_ Threshold-2 then 3 platforms are known by that employee. Example: Assuming Training Hours_ Threshold-1= 200 and Training Hours_ Threshold-2 = 500, the following platform knowledge will be determined (Table 3).
Employee ID Initial Training Hours Platforms Knowledge
1 550 3
2 230 2
3 220 2
4 220 2
5 100 1
6 90 1
7 80 1
8 70 1
9 100 1
10 100 1
Table (3) Example for the platform knowledge determination
Based on the above criteria, the Employees Experience loop is built. An initial state is fed into the model which consists of the training hours already received (Initial data) and the planned number of training hours per year based on the employees experience (imported data). In the initial trials to build the model, “the training hours “were modeled as one variable containing the total number of hours of all the employees but this didn’t work as it prevented the capability of increasing the training hours per year for each employee differently so after going through the Powersim it was found that the best way to model the training hours for each employee and the rate of increase both as arrays. An extremely powerful feature of Powersim is the possibility to define indexed variables, or arrays. One array variable can hold several values, as opposed to an ordinary scalar variable, which holds only a single value. Each array variable consists of several elements. By defining a variable as an array, a group of related values may be represented as one variable which in my case is the training hours received by different employees which gives the ability to simulate the real case which is different employees getting different training sessions . The next challenge was how to import the initial data into the Powersim model, in the beginning a manual method was used to fill the array but it was not accurate or convenient in addition that It leads to a static model i.e. cannot be changed which contradicts with the purpose to have a dynamic & flexible model and also having the data imported manually, limits the number of entries i.e. the employees to be analyzed which will yield to another drawback. Going through Powersim, a very useful function called XLDATA was found. The XLDATA function returns the values of an area in an Excel worksheet as a scalar, vector, a two-dimensional array, or a three-dimensional array. XLDATA cannot be used in composite expressions; i.e., it must define a variable completely, and it cannot be used for writing data to Excel. For example to import the following data range in excel, the XLDATA definition will be XLDATA("C:/../Book1.xlsx", "Sheet1", "R1C1:R5C2") (table 4).         Table (4) EXCELDATA example So in order to create a dynamic model as much as possible and easy to use, XLDATA function was used to create the following 2 arrays; the initial data array that contains the training hours already received for each employee so its dimension is 1 * Number of employees. The imported data array that contains the planned number of training hours per year for each employee so its dimension is 1 * Number of employees. Through this study, the employees of a given organization will be classified into 3 categories as shown in ( table 5) below;
Employee category Platforms Knowledge Representation in the array
Normal 1 platform 1
Experienced 2 platform 2
Rare 3 platform 3
  Table (5) employees Categories Using Powersim simulation, the number of training hours per employee for a future period of time (Quarters for example) can be estimated based on the settings of the simulation.(figure 3). The number of training hours received is simulated with a “level” which acts as a reservoir that keeps the flow (training hours / year) going into it. Figure (3) The platforms movement is simulated using multiple if condition as shown below; FOR( i='data range' | IF('Training hours'[i] >= 'Training Hours_ Threshold-1', IF ('Training hours'[i] <= 'Training Hours_ Threshold-2',2,3) ,1)) The below Graph in  figure (4) represents the RUN of the above model on 20 employees for 9 consecutive quarters .   The (table 6)  below shows the platform movements as an array over years As an example we can see in year 0 the following array: {3,3,2,2,2,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1} Which means we have 2 employees as Rare knowing 3 platforms, 3 employees as experts knowing 2 platforms and 15 Normal employees knowing only 1 platform. While in year 5 we have 3 employees as rare and 17 as experts and none normal. And the good thing that you can even determine that the 3rd employee is expert and the 4th is rare, {3,3,2,3,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2}   Using a “for loop”, an “IF condition” decides for each employee the number of platforms he possess based on the threshold training hours then using COUNTEQ function , the number of employees with a given number of platforms is determined .(figure 5) For example, the array A= {1,1,2,3} COUNTEQ (A,1) will be equal 2 Which means the number of repeated an element equal to “1” in the array is 2. Then to make the study more practical and matching the real case scenario, the churn of employees is taken into consideration and modeled by Exp. leaving per year and rare leaving per year. Fig (5) number of platform processed For a team to achieve certain network KPIs, it should have enough experienced and rare members so an experienced employees target count and rare employees target count are created in the model as user inputs to give the user the ability to adjust the model and the simulations based on the requirements and the experience of the decision maker which in this case will be one of the executives in the MNO. As mentioned earlier, the promotion of an employee from a level to another depends on the availability of open position(s) in the higher one so there will be cases when there are employees qualified from a knowledge perspective to be promoted but still no availability/need to promote them so to simulate this case, variables called Experience ready and Rare ready are created as these values will be needed in the decision making of promoting the employees later on.(figure 6) 8.2.2 Employees’ movement The second major part of the “employee experience” loop in the model is the employees’ movement loop which builds on the output of the platforms movement loop. In this loop, the movement of employees between the 3 categories is simulated, taking into account the following parameters;
  • The availability of qualified resources to be promoted – Experienced ready & Rare ready.
  • The availability of open positions in the higher categories – Experienced count target & rare count target.
  • The churn rate.
The available open positions are determined by the level of experience needed, for example , at a given project to be achieved , 15 % of the employees need to be experienced and 5% need to be rare so based on this count targets , the model will simulate the employees’ movement to predict when this employees combination will be available. If there are no qualified resources available to be promoted, the number of experienced or rare employees will be below the target (required) count which will increase the time needed to insource the project on hand and may lead to outsourcing the project if it is time sensitive . In the initial phase of the model, the employees’ counts were modeled as an auxiliary. An auxiliary is used to combine or reformulate information. It has no standard form; it is an algebraic computation of any combination of levels, flow rates, or other auxiliaries i.e. it has no memory. In this trial, the value of the experienced count for example was calculated at a given instant (year for example) and it will be calculated again based on the equation creating it in the next simulation step which is not adequate for the case in hand where the counts of a given year will depend on the previous one so after studying the different variable types of the Powersim tool, it was found that the best way to model the Normal, Experienced, Rare employees’ counts as “levels”. The concept of levels depends on the fact that every element in feedback loops, and therefore every element in a system, is either a level or a flow. Levels are accumulations and flows represent the changes to these levels. Flows fill up or drain the levels, much as the flow of water into a bathtub fills it, and the drain at the other end (another flow) empties it. This action of flows being accumulated in levels is the cause of all dynamic behaviors in the world. So in our case, employees’ counts at different categories are levels and the decision of promotions is the flow that fills these levels and the employees’ churn is the flow that drains it. For example, as shown in the loop below (figure 7), the Fresh to Normal decision is the flow the fills the Normal count level to compensate for the N leaving per year that drains it. Figure (7) Normal employee movement The decision of promotion flow is modeled by an IF condition that includes the parameters mentioned above. For example to model the Norm to Exp Decision, I used multiple IF condition as follows; IF ('EXP Count' <'EXP Count Target' AND 'Exp Ready'  > 0 <<emp>>, IF ('Exp Ready'>=('EXP Count Target' -'EXP Count'),'EXP Count Target' -'EXP Count','Exp Ready'), 0 <<emp/year>>) Using the first condition, the current count of the experienced employees is compared to the target count and also the availability of ready employees to be promoted is checked. Once the current count is below the target and there is availability of ready employees, the count is increased by the delta between the target and current counts on condition that this delta count is equal to the ready employees to be promoted. At each  simulation run (time unit ) quarter for example , the employees’ count of a specific category is compared to the target count , if it is less , then a number of employees are promoted to achieve the target count given that they are qualified i.e. know enough platforms . The churn can happen due to different reasons, for example staying in the same position for a long time without a promotion, salary saturation etc. In this study, the churn flow (rate) is a user input percentage for the sake of simplicity. Another major technique I used to model the employees’ movement is the feedback from one variable to another. For example in the loop below, the feedback from the EXP count to the Norm to Exp  Decision is very important to have accurate results.(Figure 8) The below charts  in (figure 9) represent the RUN of the above model on 20 employees for 10 years. As an example, if we look at year 1 we can see: 15 Normal, 3 Experienced and 1 Rare. By checking the values, we will see that the model is adjusting the counts to match the input target counts, so for example in the above simulation, the target count was 4 for experienced and 3 for rare which are the numbers the model is trying to achieve. To test the outputs of this loop, input target counts are applied as below in (Figure 10); Starting from an initial state of {16, 3, 1} and checking the output of the loop, it is clear that the model managed to achieve the target counts {12, 6, 2} starting the 5th quarter.
Figure (10) Example for the loop output
Another example to test the outputs of this loop with the following inputs as target counts              ( Figure 11); Starting from an initial state of {16, 3, 1} and checking the output of the loop, the model is approaching the target count {14, 3, 3} as it achieves {15, 2, 3} starting the 5th quarter.
Figure (11) Example for the loop output
8.2.3 Insourcing cost: Given the counts of employees at different categories, the insourcing cost can be estimated. The following parameters are taken into consideration to estimate the overall cost of insourcing (Figure 12,15);
  • Tools for the staff.
  • Transportation.
  • Training
  • Location
-Salaries including annual raise and bonuses. Cost of tools, transportation, training and location are modeled as constants which represent information that is not changed by the simulation, but they can be changed by the user through input controls based on the user requirements. Figure (12) Insourcing Cost parameters   Constants are often used to identify and quantify the boundaries of the model, and to represent decision parameters. They are, as the name implies, constant, and the definition only defines the initial value (the definition is only calculated at the start of the simulation). Also it is possible to assign a new value to a constant through input controls, thereby changing the scenario of the model. By creating permanent constants, I can create constants that not only keep their values over one simulation run, but also keep its value between simulation runs. Permanent constants help to create simulations that "remember" the input given by the user. It is useful to create constant variables rather than including literal constants in various variable definitions, this help to clean up the model and visualize parameters that might be decision parameters in the system. It also helps to gain full effect of Powersim powerful unit detection capabilities. Also, if it is needed to change units at a later stage; I will only have to do so for a handful of constants rather than going through all the variables of your system to find them. Cost of salaries is modeled as a level as it represents states in the system that change over time. Levels are variables with memory, and their value is determined by the flows that flow in and out of them. The rate of change of the salary (in-flow) is the annual raise and to take into consideration the salary saturation after a certain number of years, the annual raise of the salaries is only applied for a given number of years only such that the model provides more practical results as shown in the time graph below (Figure 13,14).
Figure (13) Employees’ Salaries over the years
As shown above, the salary saturates after year 3, which is one of the main reasons of employees’ churn. ins
Figure (14) Example for the loop output
IN SUMMARY, The researcher was able to build and test the first two loops of his model, mainly the experienced resources and the cost . The researcher is confident as per Sterman to follow the same approach to build and test the 3rd loop of the model  representing the Key Performance Indicators in a straight forward way and with the same concepts used to create the first 2 loops already explained.
  In this pilot project, I achieved wo objectives : The first was to  identify why many outsourcing agreements failed to achieve part or whole of its goals and in doing so was there a way of minimising the risk to a company if  it’s outsourcing strategy failed.  My solution was to build a decision support system. I chose the System Dynamic methodology and the modelling tool Powersim as my research showed that this was a powerful way of building such systems. The second objective of this pilot project was then to test if Powersim could do all the tasks that would be needed to create an efficient and reliable Decision Support Tool. Both objectives have been achieved. I have shown to my satisfaction that a decision support tool would be a beneficial tool for the industry and I have tested out some initial loops and functions of Powersim that I will need in my final model My tests involved modelling the insourcing which helps to predict and simulate the availability of the in-house resources and also the insourcing cost model was explained. I have also shown that  the insourcing strategy will be able to achieve certain network Key performance indicators (KPIs) and that Powersim has all the functional ability to model the next stage which is  how to model the KPIs and relate them to the available experience of the in-house staff. I am now confident that I can create  a top level design will relate the outputs of the insourcing model and the insourcing cost model to the new KPI loop that will be created. References Agency Theory as Logan, Mary S (2000)   Mary, S. L. (2000), Using Agency Theory to Design Successful Outsourcing Relationship, The International Journal of Logistics Management,Vol. 11 Iss. 2, pp. 21-32 Aron, R. and Singh, J. V. (2005). 'Getting offshoring right'. Harvard Business Review, 83, 135-43. Barney, J. B. 1991. ‘Firm Resources and Sustained Competitive Advantage.’Journal ofManagement 17 (1): 99–120. Barzel 1989   Barzel, Y. 1989. Economic analysis of property rights: Political Economy of Institutions and Decisions series; Cambridge; New York and Melbourne; Cambridge University Press. Blaxter, et al. (1996:122   Blaxter, L., Hughes, C. and Tight, M. (1996, 2002, 2006, etc.). How to Research. Buckingham, Philadelphia: Open University Press. 263 pp. Blaxter, Hughes & Tight, 1996:121 Coates and McDermott (2002)  Coates, T.T., McDermott, C.M., 2002. An exploratory analysis of new competencies: a resource based view perspective. Journal of Operations Management 20, 435–450 Crandall, R. W. / Eisenach, J. A. / Litan, R. E. (2009) ‘Vertical separation of telecommunication networks: Evidence from five countries’, available online at: /abstract=1471960  [last accessed Jan 5, 2012]. Edoardo Mollona & Alessandro Sposito (2008),  Mollona, Edoardo, and Alessandro Sposito. 2007. “Transaction Costs and Outsourcing Dynamics : A System Dynamics Approach.” International Conference on System Dynamics: 1-16. Eisenhardt, K. M. (1989), "Building theories from case research", Academy of Management Review, Vol.14, No.4, pp. 532-550. Ellram, L. M., Tateb, W. L., & Billington, C. (2008). Offshore outsourcing of professional services: A transaction cost economics perspective. Journal of Operations Management, 26(2), 148–163. Friedrich, R. / Weichsel, P. / Miles, J. / Rajvanshi, A. (2009) ‘Outsourcing Network Operations - Maximizing the Potential’, Booz & Company, Hecker, A / Kretschmer, T (2010) ‘Outsourcing Decisions: The Effect of Scale Economies and Market Structure’, Strategic Organization 8 (2): 155–175. Jacobides, M.G. and Winter, S.G.(2005)  “The co-evolution of Capabilities and Transaction Costs: Explaining the Institutional Structure of Production”, Strategic Management Journal. 26 (5): 395-413. Jiang, B. / Belohlav, J. A. / Young, S. T. (2007) ‘Outsourcing Impact on Manufacturing Firms’ Value: Evidence from Japan’, Journal of Operations Management 25 (4): 885-900. Klein et al. 1978  Klein, B., Crawford, R. and Alchian, A. (1978), Vertical Integration, Appropriable Rents and the Competitive Contracting Process, Journal of Law and Economics, No. 21, pp. 297-326 Lowson, R. H. (2002). Assessing the operational cost of offshore sourcing strategies. International Journal of Logistics Management, 13(2), 79–89. "Madhok, A. (2002). Reassessing the Fundamentals and Beyond: Ronald Coase, the Transaction Cost and Resource-Based Theories of the Firm and the Institutional Structure of production, rategic Management Journal, 23, pp.535-550." Marshall, D., McIvor, R. and Lamming, R. (2007) Influences and outcomes of outsourcing: insights from the telecommunications industry. Journal of Purchasing and Supply Management 13, 245-260. McNally and Griffin 2004  McNally, R. C., & Griffin, A. (2004). Firm and individual choice drivers in make-or-buy decisions: a diminishing role for transaction cost economics?. The Journal of Supply Chain Management, 40(1), 4–17.  Monteverde, K. (1995). Technical dialog as an incentive for vertical integration in the semiconductor industry. Management Science, 41(10), 1624–1638. Poppo, L. / Zenger, T. (1998) ‘testing alternative theories of the firm: transaction cost, knowledgebase, and measurement explanations for make-or-buy decisions in information services’, Strategic Management Journal 19 (9): 853-877. Teijlingen, E. R. and Hundley, V. (2001), The Importance of Pilot Studies, Social Research Update, Issue 35, University Vastag, G., 2000. The theory of performance frontiers. Journal of Operations Management 18 (3), 353-360. Williamson, O. E. (1991) ‘Strategizing, economizing, and economic organization’, Strategic.  Management Journal 12: 74-94. Youngdahl, William; Ramaswamy, Kannan Journal of Operations Management, 2008, Vol.26(2), pp.212-221 [Peer Reviewed Journal]

Cite This Work

To export a reference to this article please select a referencing stye below:

Reference Copied to Clipboard.
Reference Copied to Clipboard.
Reference Copied to Clipboard.
Reference Copied to Clipboard.
Reference Copied to Clipboard.
Reference Copied to Clipboard.
Reference Copied to Clipboard.

Related Services

View all

DMCA / Removal Request

If you are the original writer of this dissertation and no longer wish to have your work published on the website then please: