Integral Workforce Planning

(1)

UNIVERSITY OF GRONINGEN

Integral Workforce Planning

An overview of factors that can be used in

mathematical modelling

Carolina Spronk 6/22/2015

Company names are not shown in this public version

(2)

Integral Workforce Planning 2

Colophon

Title: Integral Workforce Planning

Subtitle: An overview of factors that can be used in mathematical modelling

Principal: Randstad

Diemermere 25 1112 TC Diemen www.randstad.nl Educational institute: University of Groningen

Faculty of Economics and Business Master Supply Chain Management Post-office box 72

9700 AB Groningen www.rug.nl

Author: Carolina Spronk

s2798611

g.c.spronk@student.rug.nl carolinaspronk@gmail.com Supervisor: Prof. Dr. I. F. A. Vis

i.f.a.vis@rug.nl

Co-assessor: Prof. Dr. K. J. Roodbergen k.j.roodbergen@rug.nl External supervisor: R. Jansen

ronald.jansen@randstadgroep.nl

Location: Groningen

Date: 22 June 2015

(3)

Preface

This master's thesis has been written as a completion of the master program Supply Chain Management at the University of Groningen. This research was executed in collaboration with Randstad, an employment agency in the Netherlands. The core of this research concerns the key factors that are important in integral workforce planning in the manufacturing and logistics sector. During the process of writing, several people contributed to complete this thesis. Therefore, I would like to take this opportunity to express my gratitude to them. First of all, I would like to thank my supervisor Prof. Dr. Iris Vis for allowing me to conduct my thesis under her supervision. Thank you for your support during my research and all you have taught me. You have helped me not only in writing this thesis but also in increasing my self-confidence by expressing confidence in my capabilities. I really enjoyed working under your supervision and I am grateful for all you did for me.

Furthermore, my thanks goes to Paul Buijs who has given me critical and valuable feedback on my thesis. He has helped me to look at my thesis from a different point of view, which has led to a better report. Thanks to Prof. Dr. K.J. Roodbergen for co-assessing my thesis.

In addition, I would like to thank Ronald Jansen, concept manager at Randstad, for all his help. As my external supervisor he has helped me by providing information on integral workforce planning, cases and a list of respondents. Furthermore, he has helped me by reading my report and giving valuable feedback. He was always willing to take the time to help me. I also would like to thank all interviewees and respondents of the different companies for their time. Without all their input I would not have been able to conduct my research.

Finally I would like to thank my family and friends for their interest in my research as well as for their love and support. A special thanks goes to my girlfriend who has helped me in every way she could. She was always there to cheer me up. Marjon, you are the best.

(4)

Abstract

High planning uncertainties and fluctuating workforce demand in supply chains are the result of the fact that companies are searching for ways to improve their operations to remain competitive. Therefore, it is essential to have an efficient and effective process of workforce planning. An option to increase the effectiveness and efficiency of this process is outsourcing. However, literature states that there is hardly any development in outsourcing workforce planning as companies are reluctant to give this process out of hand. Especially regarding the planning of permanent employees. As there is a lack of research on outsourcing workforce planning, this study investigates key factors that are important in integral workforce planning (IWP). IWP is defined as making workforce plans for both the temporary and the permanent employees in one system, conducted by an employment agency. The focus of this research was on the manufacturing and the logistics sector. A literature study was conducted as well as an embedded multiple case study and a survey. The literature study and the embedded multiple case study were used to retrieve factors that could be important in IWP. The survey provided additional insights and allowed the ranking of the factors found. Scores were allocated to the rankings on which the final ranking was based. The following four key factors were identified:

1. The amount of employees needed (both permanent and temporary employees); 2. The qualifications of the employees;

3. The availability of the employees;

4. Flexibility by means of multiple qualifications per employee.

(5)

List of figures

Figure 1. Methodological framework ... 12

Figure 2. Interview protocol ... 17

Figure 3. Characteristics of the respondents of the survey ... 31

Figure 4. Mathematical model of Wirojanagud et al. (2007) ... 39

List of tables

Table 1. Journals used for the literature study ... 10

Table 2. Case selection ... 10

Table 3. Factors that are important in workforce planning based on literature ... 16

Table 4. Factors that are important in IWP based on practical study at Company A ... 19

Table 5. Factors that are important in IWP based on practical study at Company B ... 21

Table 6. Factors that are important in IWP based on practical study at Company C ... 24

Table 7. Factors that were found in all three cases studied ... 27

Table 8. Comparison between the manufacturing sector and the logistics sector ... 28

Table 9. Overview of the factors found based on literature study and practical insights ... 28

Table 10. List of costs that was used in the survey ... 30

Table 11. List of factors that is used in the survey ... 31

Table 12. Results on the list of costs ... 32

Table 13. Results on the list of factors ... 33

Table 14. Ranking of the factors ... 33

(8)

1. Introduction

Companies are continuously searching for ways to improve logistics operations to remain competitive in the current world of tight delivery schedules and high service levels. These trends have resulted in high planning uncertainties and fluctuating workforce demand in supply chains. Labour is often a large cost component (Eveborn & Rönnqvist, 2004). For that reason, workforce planning is the critical missing component of performance improvement and strategic management initiatives (Cotten, 2007). It is therefore important to keep this process efficient and effective. One of the main advantages of outsourcing is to make the outsourced process more effective and efficient (Lankford & Parsa, 1999). Being effective and efficient is important for workforce planning, and it is therefore relevant to study the factors that are important in outsourcing workforce planning. Another reason why this is interesting is the fact that the attention for the Physical Internet is increasing. Its vision aims at addressing the current unsustainable way we transport, handle, store, realize, supply and use physical objects around the world by using the Internet as an underlying metaphor for steering innovation in the physical sphere (Montreuil, 2011). This would indicate an increase in the sharing of, among other things, trucks and employees in a shared transportation and distribution network. Hence, the amount of flexible employees will increase, making it essential to have a kind of director who can oversee everything.

1.1 Problem definition

Workforce planning problems have been widely studied in the last few decades (Van den Bergh, Beliën, De Bruecker, Demeulemeester, & De Boeck, 2013), especially due to its complexity, its challenging features, and its practical relevance (Brucker, Qu, & Burke, 2011). Nowadays, there are many mathematical models that can be used to make staffing schedules. Van den Bergh et al. (2013) presented a literature review on this subject. However, there is a relatively low level of study on outsourcing this process. Previous research on workforce planning has mainly focused on in-house planning and has not taken into account the trends that are described above. As mentioned before, instead of making workforce plans in-house, companies can also outsource staff planning decisions to employment agencies. In addition, Jansen and Smit (2013) state that there is hardly any development in outsourcing workforce planning as companies are reluctant to give this process out of hands. Outsourcing workforce planning requires a different approach and modelling, especially when a director oversees multiple companies. Therefore, the key factors that are important in modelling outsourced workforce planning will be investigated. There are advantages to outsourcing total workforce planning, which can be measured using several performance measures. For example, at Company X, the overall equipment effectiveness improved by 5% on average (Randstad Inhouse Services, 2014). Therefore it is important to know which performance measures are connected to the factors.

(9)

Integral Workforce Planning 9 A company can opt for full or partial outsourcing. For example, the company can choose to only outsource workforce planning for flexible employees. This process is very common and is therefore efficient while workforce planning for fixed employees is still handled internally. However, integral workforce planning for both fixed and flexible staff can lead to additional advantages (Randstad Inhouse Services, 2014). For example, with integral staff planning concepts employment agencies can assist in determining an optimal mix of in-house and externally hired staff. In this study, integral workforce planning (IWP) is defined as making workforce plans for both temporary and permanent employees in one system, conducted by an external party (employment agency). As different parties are involved in outsourcing, different views are important to investigate. Factors that are important in IWP can differ per view. The outsourcing companies may include other aspects than the employment agency. As different views exist, it is interesting to study the possible differences in comparing the parties involved in IWP with the parties that did not implement IWP and the factors stated in literature. Additionally, a comparison will be made between the manufacturing and the logistics sector to give first insights in possible differences between these sectors.

The goal of this research is to provide a list of factors that are important in IWP by means of a literature review and an embedded multiple case study among several Dutch companies in the manufacturing and logistics industry within the context of an employment agency. Semi-structured interviews and a survey are conducted to gain data upon which the findings are based. Factors used in mathematical models in standard literature will be retrieved and compared with the practical insights. A final list of key factors will be provided which can be used in further research in a mathematical model to optimise IWP.

1.2 Research questions

As it is important to know the overall factors that influence IWP, the main research question of this research is: "What are key factors that are important in integral workforce planning?" This research question can be divided into several sub-questions as stated below:

1. What are factors that are important in IWP based on literature?

2. What are factors that are important in IWP based on practical insights according to: a. employment agencies?

b. companies in the manufacturing sector? c. companies in the logistics sector?

3. What are the performance measures that are connected to the factors found?

4. What are the differences in factors between the manufacturing and the logistics sector?

5. Which of the factors are most important to include in a mathematical model? (What is the ranking of the factors found?)

6. How can the key factors be included in a mathematical model?

1.3 Methodology

Since there are various research questions in this study, different approaches are used to collect the needed data. This section will represent the methods used.

1.3.1 Literature review

(10)

Integral Workforce Planning 10 Knowledge and Google Scholar were used to collect the articles that are used for the review. Examples of search terms used are 'workforce planning review', 'integral planning factors', and 'staff planning state-of-the-art'. Articles from top journals in operations research and more recent than the year 2000 were given preference. An overview of the journals used can be found in table 1.

Table 1. Journals used for the literature study

Journals used Frequency

European Journal of Operational Research 6

Annals of Operations Research 4

Omega 2

Operational Research Quarterly 1

International Journal of Production Research 1

1.3.2 Practical insights

To gain practical insights and to answer sub-questions two to four, an embedded multiple case study has been conducted. The context that is used in this research is that of the employment agency Randstad. Randstad is working in both the manufacturing and the logistics sector, which makes this company suitable for this study. In order to be able to compare factors that are important in IWP from the view of the employment agency to the factors based on the view of the outsourcing companies, multiple cases have been selected. This approach was chosen as rich data can be ensued from semi-structured interviews, informal information that is gathered by visiting the company, and from having face-to-face meetings with representatives of the company. Contradictory to a survey, a case study is unconstrained by the rigid limits of questionnaires and models and it can lead to new and creative insights and the development of new theory (Voss, Tsikriktsis, & Frohlich, 2002), which is a good fit for this research as the concept of IWP is new. However, it should be noted that care is needed in drawing generalisable conclusions from a limited set of cases.

It was decided to focus on multiple companies as multiple case studies typically provide a stronger base for theory building and exploration (Voss, Tsikriktsis, & Frohlich, 2002). It allows for comparisons that clarify whether an emergent finding is simply idiosyncratic to a single case or consistently replicated by several cases (Eisenhardt, 1991). Further, Eisenhardt (1991) states that insights from multiple case studies are better grounded, more accurate and theoretically transferable when compared to a single case study design. The companies are all located in the Netherlands and are customers of the employment agency Randstad.

In selecting the cases, literal replication logic was applied to increase the external validaty. In literal replication, similar cases, similarities in outcomes can be explained by similarities in variables. Most similar and most different cases were selected. The most important variables where the case selection was based on were the sector and whether or not integral workforce planning is already implemented. The case selection is summarised in table 2.

Table 2. Case selection

Case 1 Case 2 Case 3

Sector Manufacturing Logistics Logistics

Already implemented IWP Yes Yes No

(11)

Integral Workforce Planning 11 researcher. The interviewees were high ranked professionals such as a senior process manager. It was chosen to conduct semi-structured interviews as they are well suited for the exploration of perceptions and opinions of respondents regarding complex and sometimes sensitive issues. They also enable probing for more information and clarification of answers (Barriball & While, 1994). Unclear answers were clarified through the use of follow-up questions. The interviews were held on-site as well as via Google Hangouts in order to save time.

Second, for the multiple case study, the companies were visited once by one researcher conducting the data collection. Again, semi-structured interviews were held taking one to two hours. The interviewees were the ones that are making the workforce planning within the companies, and higher ranked professionals that are involved in the workforce planning process such as an operations manager. The interviews were audio recorded by digital devices in case of granted permission. Further, the researcher took notes during the interviews. These recordings and notes helped the researcher to accurately summarise the relevant responses made by the interviewees. To execute the interviews, an interview protocol was used. This will be explained further in chapter 3. And third, during the site visits, factory tours were used as an opportunity to ask further questions. Additionally, the process of making workforce plans was observed and the researcher participated in the planning process. This way more insight was gained in this process.

In case studies there are two steps in data analysis: analysis within case data and cross-case analysis (Eisenardt, 1989). Regarding the analysis within the cases, the researcher researched the different codes to gain insights in the underlying relations. Relevant quotes were given themes (descriptive codes). Secondly, the descriptive codes were combined. These combinations were provided with other labels, which can be regarded as categories (interpretive codes). The labels and codes were partly retrieved from the factors found in literature. Quotes that could not be coupled to these factors, were given a new label. To compare the results of the different cases, a cross-case analysis is needed. The cases were compared in pairs. First, this was done for one pair of most similar cases. This means that the cases of the logistics sector will be compared to each other. After that, the most different cases were compared. In this comparison, the two sectors are compared to each other. By comparing the less likely cases, it became clear whether discrepancies exist between the different cases.

1.3.3 Validation and ranking of the factors found

In order to validate the factors found and to state which of these factors are most important given certain performance measures, a survey was handed out to several companies in the manufacturing and logistics sector. This method was chosen because one of the main strengths of the survey method is that it has generalisation capacity (Karlsson, 2009). With this survey additional insights were retrieved as well as a ranking of the found factors from the case study. The survey will be further explained after sub-questions one to four are answered.

1.3.4 Prelude to a mathematical model

(12)

Integral Workforce Planning 12 The research design can be summarised as shown in figure 1.

Figure 1. Methodological framework

1.4 Outline of this paper

The literature review is presented in chapter two. In this chapter the first sub-question will be answered. In chapter three, the answers to sub-questions two to four can be found: the practical insights. Chapter four covers the survey: the factors found in chapters two and three will be validated by means of a ranking (sub-question five). Chapter five contains the discussion. In this discussion the results of this research will be discussed. Additionally, a prelude to a mathematical model will be given to answer sub-question six. The final chapter, chapter six, will describe the conclusion and possibilities for further research.

Factors that are important in integral workforce planning based on literature review

Factors that are important in integral workforce planning based on practical insights

List(s) of factors that are important in integral workforce planning in both the manufacturing and the logistics sector that can be used in a mathematical model

(13)

2. Literature review

High planning uncertainties and fluctuating workforce demand in supply chains are the results of the fact that companies are continuously searching for ways to improve logistics operations to remain competitive. Cotton (2007) states that workforce planning is the critical missing component of performance improvement and strategic management initiatives. Due to current trends and the criticality of the process it is important to keep staff planning efficient and effective. One way of achieving this is to outsource the total workforce planning to an employment agency.

In order to fully understand the problem, first the definition of outsourcing needs to be clear. Outsourcing has been defined in many ways. For example, Lankford and Parsa (1999) define outsourcing as "the procurement of products or services that are external to the organization" and Laabs (1993) defines it as "having an outside vendor provide a service that you usually perform in-house". The essence of the definitions is that outsourcing refers to the concept of looking for expertise to handle certain business functions outside the existing firm (Embleton & Wright, 1998). Lankford and Parsa (1999) argue that firms should consider outsourcing when it is believed that certain support functions can be completed faster, cheaper, or better by an outside organization. The goal of integral workforce planning is to increase the efficiency and effectiveness of this process and to simultaneously increase the productivity of the outsourcing company.

The goal of this study is to provide a list of factors that are important in IWP and can be used as input for a mathematical model. A mathematical model consists of several components. Decision variables are the controllable aspects of a problem. The aspects that measure how well the objective of the decision maker are achieved form the performance measure. The uncontrollable inputs are referred to as parameters (Daellenbach, McNickle, & Dye, 2012). The performance measure is the component that the decision maker wants to maximise or minimise. It is the basis of the model. It is therefore important to identify the performance measure. The decision variables and parameters are used to achieve the objective. So, factors can be used as decision variables, performance measures and parameters.

Different factors play a role in IWP. In order to gain insights in possible influencing factors, a broader scope is taken into account. Not only integral workforce planning concepts in literature have been studied, but also workforce planning for both permanent and flexible staff separately as IWP is a combination of these two. Factors found in previous studies will be discussed in this chapter. Table 3 contains an overview of these factors.

2.1 Amount of employees

(14)

2.2 Level of skills of the employees

One set of factors that can be derived from literature is about the level of skills of the employees. Among others, Maenhout and Vanhoucke (2013) and Fowler, Wirojanagud and Gel (2008) integrated the skill set of a worker in their model. Fowler, Wirojanagud and Gel (2008) define the set of skills as the set of machine groups the employee can operate. If the level of skills is higher rather than lower, less employees are needed in order to operate the machines, which will therefore have an influence on workforce planning. De Bruecker, Van den Bergh, Beliën, and Demeulemeester (2015) studied the influence of the level of skill of the employees on workforce planning. In their study the authors identified six different skill determinants. These determinants are the age and the experience of the employee, the degree of technical knowledge, the licenses, qualifications and job title he received, and the nurse grade (in health care applications) of the employee. Not only the level of skill itself is important, but also the requirement per task for different competency skills (Maenhout & Vanhoucke, 2013).

2.3 Costs

Another set of factors that has an influence on workforce planning according to literature is about different types of costs. Van den Bergh et al. (2013) mention the following subdivision: personnel cost, different cost per day, different cost per skill, overtime cost, outsource cost, travel cost, and cost of executing tasks. Regarding the different cost per day, one should take into account that there are increased wages for weekend days. As mentioned in section 2.1, personnel cost has a wide range of possibilities. This range of possibilities contributes to the popularity of assigning cost to factors compared to the minimisation of the number of employees (Van den Bergh et al., 2013). The cost for the employees that are assigned is also mentioned in other studies (e.g. Brucker, Qu, & Burke, 2011; Wirojanagud, Gel, Fowler, & Cardy, 2007). In addition to the previous mentioned cost, hiring and firing cost should be taken into account (e.g. Fowler, Wirojanagud, & Gel, 2008; Li, Chen, & Cai, 2007). Other costs are for example, cost of training (De Bruecker et al., 2015), cost of missed production (Wirojanagud et al., 2007), etc.

2.4 Time-related factors

(15)

2.5 Under- and overstaffing

Under- and overstaffing are also factors that can influence workforce planning. Workforce planners should consider whether or not these factors are allowed. If it is allowed, it means that the workforce demand does not have to be met exactly. In contrast, one can schedule some excess staff to deal with, for example, unexpected demand (Van den Bergh et al., 2013). Meanhout and Vanhoucke (2013) point out that penalties could be received when under- or overstaffing occurs in case it is not allowed. Fowler, Wirojanagud and Gel (2008) also incorporate overstaffing in their heuristic. It is important to keep under- and overstaffing in mind as it can lead to additional costs. In addition, Mani, Kesavan and Swaminathan (2014) found that understaffing leads to a decline in employee satisfaction and therewith in the financial performance of the company.

2.6 Preferences

Next to all previous mentioned factors, preferences of employees is also a factor that can be taken into account. For example, Knust and Schumacher (2011) indicate preference lists of acceptable drivers for every truck. They also state that many companies want to use their staff in an efficient way trying to take into account preferences of their employees as much as possible (Knust & Schumacher, 2011). An example is the research of Rasmussen, Justesen, Dohn and Larsen (2012) who sought to minimise the number of uncovered visits and maximise the preference level of the solution. Ernst, Jiang and Krishnamoorthy (2004) also mention that preferences of employees often is one of the constraints of workforce planning. These preferences can have an influence on the costs, as it might be more efficient not to look at preferences. It therefore depends on the performance measure of the company whether or not to take preferences into account. Workforce planning with employee preference constraints are most common in the health care sector but can also be found in other sectors (Van den Bergh et al., 2013).

2.7 Other related factors

Another factor that is related to workforce planning is, of course, the demand. The workforce planning is made in order to meet demand and therefore this factor often plays an important role (e.g. Eveborn & Rönnqvist, 2004). In addition, flexibility (e.g. Brucker, Qu, & Burke, 2011) and breaks (Ásgeirsson, 2014) are influencing factors. Next to that, factors relating to shifts can influence workforce planning. Examples are the sequence of the shifts and whether or not shifts are able to overlap (e.g. Knust & Schumacher, 2011). Another set of factors is related to the tasks. As tasks can differ, there is also a difference in workload. Additionally the speed of work should be taken into account (Fowler, Wirojanagud, & Gel, 2008). According to Eveborn and Rönnqvist (2004) task rotation also influences workforce planning. Among others, Wirojanagud et al. (2007) mention the importance of including training. For example, during a period of training, loss of labour occurs, which matters for workforce planning. Lastly, Wruck (2013) provide an analysis of risk modelling approaches for warehouse staffing problems.

2.8 Overview of the factors

(16)

Table 3. Factors that are important in workforce planning based on literature

Factors 1 2 3 4 5 6 7 8 9 10 11 12 # employees needed (permanent/temporary) X X X X X X X X X Skills employees X X X X X X Costs Personnel cost X X X X X X X X

Different cost per skill X X Overtime cost X X X Outsource/casual workers cost X X X Hiring/firing cost X X X Missed production cost X X X Other X X X X Tasks X X X X X X Task rotation X Speed of work X Flexibility X X (Cross)Training X X X X Shift sequence X X X X Shifts overlap X X X Under/overstaffing X X X X X X Breaks X Preference X X X X X Time-related

Min & max # assignments

X X X X X

Min & max # consecutive days

X X X X X X X

Min & max # consecutive days off

X X X X X

Min & max # hours X X X X X X X X X X

Max # overtime X X X X

Time between assignments

X X

Free days after night shift X X X Time horizon X X X X X X X Holidays X X Other X Demand X X X X X X X X Risk X

1. (Maenhout & Vanhoucke, 2013) 2. (Brucker, Qu, & Burke, 2011) 3. (Fowler, Wirojanagud, & Gel, 2008) 4. (Wirojanagud, Gel, Fowler, & Cardy, 2007) 5. (Li, Chen, & Cai, 2007)

6. (Wruck, 2013)

7. (Eveborn & Rönnqvist, 2004) 8. (Baker, 1976)

9. (Azmat, Hürlimann, & Widmer, 2004) 10. (Ásgeirsson, 2014)

11. (Knust & Schumacher, 2011)

(17)

3. Practical insights

In order to gain practical insights, an embedded multiple case study has been conducted as explained in section 1.3.2. This chapter will discuss the case study.

3.1 Interview protocol

As mentioned in section 1.3, interviews were held on site of three different companies. An interview protocol was used to guide the interviews. Two different protocols were used as a division was made between companies that already implemented integral workforce planning and companies that have not. These protocols can be found in Appendix 1 and 2 and will be explained below.

The interview protocols are divided into four parts. The first part contains questions regarding the function and the role of the interviewee within the company. The second part is about the perspective of the company regarding (integral) workforce planning. For example, it was asked why it was decided to introduce IWP and how the employees responded to this change. In the third part, questions are asked regarding the planning process itself. Here, the interviewee is asked to explain the planning process and the factors that play a role in this process. Subsequently, questions are linked back to the theory by contributing to factors found in literature (see table 3) that have not been mentioned by the interviewee. For example, if the interviewee did not say anything about preferences (section 2.6), it was asked specifically whether or not this factor plays a role in their workforce planning. In the last part, the interviewees are asked to give the three most important factors. A summarising representation of the interview protocols can be found in figure 2. As different employees with different functions will be interviewed, specific questions that were asked were different per function. For example, the higher ranked professionals that are involved in the workforce planning process were asked questions that are more in depth about the company's perspective regarding (integral) workforce planning whereas the planners were asked questions that are more in depth about the planning process itself.

Figure 2. Interview protocol

A. Background interviewee

 Function and role

 Experience and time-span of function

B. Company perspective regarding (integral) workforce planning

 Strategy regarding to workforce planning

 Reasons (not) to introduce IWP

 Response of employees to the implementation of IWP

C. Planning process

 Planning process

 Factors that influence the planning process

 Mention factors from table 2 (section 2.8) that have not been mentioned by the interviewee

D. Top 3 factors

(18)

3.2 Case study

As mentioned in section 1.3, the embedded multiple case study is divided into two sectors. Within the manufacturing sector one case has been studied and within the logistics sector two companies have been visited. This section will describe the case study performed.

3.2.1 Manufacturing sector

The case of Company A has been studied as positioned in the manufacturing sector. Company A is the leader in lighting, driving the digital lighting revolution. They are using customer insights and technological innovations, coupled with their trusted brand and global leadership positions, to lead the digital lighting revolution (Source not shown as it contains confidential information). Due to the innovation there are a lot of fluctuations. Therefore, the company is working with projects and different scenarios. The production of the company is divided into units that make different products. Company A has recently implemented IWP and therefore Randstad is making their total workforce planning. However, not every unit is included in this process. Integrating all units into IWP is a plan for the near future.

In an interview with an in-house employee of Randstad, it became clear that the workforce plans at Company A are based on demand and forecasts. As the production is fluctuating, the company makes use of scenarios. Within the process of planning, the qualifications and skills of the employees are most important. First, the fixed employees are planned. Every fixed employee has a home base. This indicates the production unit for which they have the highest qualifications. Based on their qualifications, the employees are assigned to a certain task. In case of more qualifications, the employees can be assigned to units other than their home base. Additionally, the in-house employees of Randstad have the ability to look at the qualifications that are available. Therefore they can advise Company A in training their employees in order to get more qualifications. This way, (cross) training is taken into account in workforce planning as well.

Time-related factors are taken into account by means of the system of Randstad. The restrictions of the law on working hours are automatically taken into account in this system. Furthermore, in case of days off and holidays the in-house employees of Randstad give an advice to the company. However, the unit leaders have the final say in this aspect. Randstad can recommend not to give an employee a day off, but if the company decides to do so, they have to find a replacement. Besides these factors, the availability of the employee is important. If an employee is not available, the system will not show this employee and therefore this employee cannot be scheduled.

It was mentioned by the in-house planner of Randstad that costs are important in workforce planning. For example, travel cost and personnel cost are considered. However, these cost-related factors are considered beforehand. This means that these costs are defined in the pool of employees and are not considered in the planning process itself. Flexibility is taken into account in the division of tasks. In case of multiple qualifications of an employee, we can be flexible in scheduling this employee. Furthermore, the amount of employees can differ per day. In order to be flexible a pool of temporary employees is available to quickly find available employees, which also allows for compensating risks.

(19)

Integral Workforce Planning 19 These preferences are not taken into account anymore in IWP by Randstad. However, if Company A wants it to be taken into account this would be possible.

A summary of the factors found Company A can be found in table 4. The quotes in this table are based on the interviews, the factory tour and based on questions that were asked during observing and participating in the planning process.

Table 4. Factors that are important in IWP based on practical study at Company A

Factors Quotes

# employees needed (permanent/temporary)

"The amount of fixed and flexible employees are put in the same system. I get a paper that says how many people are needed including the scenario that goes with it. Then my job is to first assign the permanent workers and afterwards I can assign temporary workers."

Qualifications of employees "We plan according based on the qualifications of employees. This is the most

important aspect in making a schedule."

Availability of employees "We can only plan employees that are available. When they are not available, the employees are not shown in the system and we therefore cannot schedule them."

Costs

Travel cost "It is not an option to save on travel cost."

Personnel cost "The personnel cost are taken into account in the process of making the pool of employees. So it is a factor that is considered. However it is taken into account in advance and not in the planning process itself."

Tasks "For each task certain qualifications are required. It is therefore important to

look at a task and see who has the skills to work there. However, every task has to be fulfilled and we therefore do not take specific tasks into account. We look at the qualifications of the employees. "

Task rotation "When I am making the schedules I try to rotate the tasks if the employees

have the qualifications to do so."

Speed of work Not applicable.

Flexibility

Multiple qualifications "As I said, there is flexibility in the division of tasks. As schedules are only made

for two days, everything can change in the new schedule, as long as there are qualifications of course. When employees has multiple qualifications we can assign them to multiple tasks."

Amount of employees "The amount of employees can differ per day."

(Cross)Training "We can give Company A the advice to train people in order to get more qualifications, because we can see in advance which qualifications are available."

"Company A created a test-line which will be used in the near future. At this line, flexible employees have to train themselves for a day before they can start working in the units."

Shift sequence Not applicable.

Shift overlap Not applicable.

Under/overstaffing "In case of understaffing, temporary workers are called and asked if they can come to work."

Breaks Not applicable.

Preferences

Preferences of employees Not applicable. Preferences of unit leaders Not applicable.

"In the past preferences of unit leaders played an important part in making workforce plans. With IWP this is not taken into account anymore. However, it is possible to do so if the company requests it."

Time-related factors

Days off "When employees want a day off, we give an advice to the unit leaders. However, they can decide to go against this advice. This should be taken into account."

Law of working hours "Time-related factors are taken into account as they are automatically inserted into the system."

Time horizon "Workforce plans are made weekly. However, changes can be made daily."

(20)

Integral Workforce Planning 20 Risk "The pool of flexible workers can serve as a back-up in order to cope with risks."

3.2.2 Logistics sector

Within the logistics sector, two companies were visited. These cases will be described in this section.

3.2.2.1 Company B

The first case in the logistics sector is a depot of Company B. Company B is one of the largest providers of express delivery services in the world. Their service extends over more than 200 countries and they are particularly strong presence in Europe. They ensure that parcels, documents and pieces of freight are delivered safely and on time, all over the world. Their European road network is unrivalled and connects more than 40 European countries, 19 road hubs and 550 depots (Source not shown as it contains confidential information). One of these depots is the depot in Eindhoven which was visited for this research. Since 2010, the company is getting total workforce plans via Randstad by means of IWP.

In an interview with the senior process manager of Randstad it became clear that Company B had decided that it did not want to focus on activities that are not their core activity. Therefore, the company decided to outsource the process of workforce planning to an employment agency. Cost savings are very important for Company B and this should be taken into account when making integral workforce plans. The workforce plans are based on forecasts of the unit leaders. Company B wants their permanent employees to be able to see their schedule three weeks in advance and therefore the workforce plans are based on forecasts for these weeks. However, demand fluctuates frequently and is seasonal. Therefore, the plans can change every day. Consequently, flexible employees are used to compensate for this, but the plans for this staff can change even less than 24 hours before a shift starts.

Cost savings are not always priority in making workforce plans at Company B. In an interview with one of the in-house planners of Randstad at Company B it became clear that at the end of last year quality was more important than costs. In that period, more employees were needed and scheduled. That costs are not always priority also became clear by talking to a warehouse manager. He explained that for him costs are less important than having employees with the right qualifications and capabilities at the right tasks. Another factor that should be taken into account is the productivity of the employees. Employees that are more productive should get preference over the less productive ones. Preferences of employees and unit leaders are, according to Randstad, not taken into account. However, in an interview with the warehouse managers of Company B it became clear that his preferences are taken into account when it comes to which employee to schedule first. So, for making a workforce plan it is important to know what Company B finds most important. The in-house employee of Randstad clarified that cost savings are not realised by eliminating the amount of employees but by shortening shifts. This time-related factor shift length is also a way of being flexible.

(21)

Integral Workforce Planning 21 overview of all employees. As Randstad has the overview and the responsibility, the unit leaders can focus on their own core tasks. This has led to an increase in efficiency. However, for the unit leaders there are several aspects for which they now have less control. For example, in the past there was a pool monitor in which the managers could control the amount of employees that are in the pool. Now the managers do not have the ability to monitor this pool.

A total overview of the factors found at Company B can be found in table 5. The quotes in this table are based on the interviews and on questions that were asked during observing the planning process.

Table 5. Factors that are important in IWP based on practical study at Company B

"Company B tells us [Randstad] the amount of employees they need. Then at first we plan their fixed employees. After that we can see the amount of flexible employees that are still needed."

Qualifications of employees "The qualifications of the employees are most important in making workforce plans. When needed qualifications are not present, we [Randstad] cannot schedule this employee."

"We [Company B] have to select the qualifications that employees have, otherwise they do not show up in the system of Randstad."

Availability of employees "The availability of the employees is also very important of course. We [Randstad] cannot plan someone if that person is not available."

Productivity of employees "For us [Company B] it is very important that next to the qualifications, the employees are productive. I would rather have employees that are 'eights' and 'nines' than 'sevens'."

"The qualifications and the productivity of the employees are, for me [Company B], more important than costs."

Costs

Different cost per skill "Differences in costs are automatically taken into account in the pool. There are different phases of employees that have different costs and different skills. Tasks that require only the basic skills have to be performed by employees with the lowest phase. These differences in cost are automatically taken into account."

Training cost "Training costs and the regulations that go with it are incorporated in the pool in advance."

Tasks "There are several tasks that need different qualifications. There are certain critical tasks. We want every employee to have qualifications for one critical task. Tasks are combined into task groups for which the same qualifications are needed."

Task rotation "Within the task groups, task are rotated. This way everyone that has certain

qualifications can perform task in which those qualifications are needed. Therefore the employees stay up to date and less training is needed."

"Employees that are good at a specific critical task mostly perform that same task."

Flexibility "For us [Company B] as a company, flexibility is very important. I would say it is one of the three most important factors."

Multiple qualifications "Employees with multiple qualifications can be scheduled more flexible as they can perform more tasks."

Amount of employees "We have some flexibility in the workforce plans as we [Randstad] plan one to two employees extra every day."

"The schedules for the flexible staff can change until the last moment, so we have some flexibility here."

(Cross)Training "While planning, we can see a couple weeks ahead. Therefore we have overview at the qualifications that are needed. (Cross)Training is taken into account this way."

Shifts overlap Not applicable.

Under/overstaffing Not applicable.

(22)

Preferences

Preferences of employees Not applicable.

"Preferences of employees or preferences of Company B are not taken into account. However, this would be possible if the company wants it."

Preferences of unit leaders Not applicable.

"Preferences of team managers are not taken into account. For us it is important to not only schedule employee one but also schedule employee two. Otherwise employee two will leave soon and then we have to find someone new."

"If I [Company B] find that someone is good I want Randstad to schedule him first after scheduling the fixed employees. They do take this into consideration if I want them to."

Time-related factors "Time-related factors are automatically taken into account in the system."

Law of working hours The so-called Arboportaal is included in our system. If according to this law of working hours someone cannot work, this employee cannot be selected in the system."

Shift length "There are many different tasks and different small shifts. These shifts differ in time."

"The shift length is flexible. The employees are sent home when the work is done."

"In order to save costs, the shifts per employee are shortened."

Time horizon "The workforce plans for the fixed employees have to be ready three weeks ahead."

Demand and forecasts "We [Randstad] plan based on forecasts of demand. We have to make the plans three weeks ahead and therefore we need forecasts to make the workforce plans."

"There is a basic planning per week. We [Company B] send this basic planning to Randstad. When we see that a period will be busier we add an extra employee to this basic planning. These periods can for example be seen in advance by looking at numbers of previous years."

Risks "In order to cope with a little risks we have some reserves. These employees can be called if needed. This also makes us more flexible."

Communication "A factor that is important is that communication between Randstad and the permanent staff proceeds via the unit leaders. When someone is sick, this is reported to the unit leader who then communicates with Randstad that an extra employee is needed."

"Generally, there is no contact between Randstad and the fixed employees. We [Randstad] only communicate with the unit leaders."

3.2.2.2 Company C

The second case in the logistics sector is Company C. Company C Logistics is a partner in international transport and logistics. They carry goods quickly and efficiently within Europe by truck or train or worldwide by ship or plane (Source not shown as it contains confidential information). Currently Company C still makes their own workforce plans for their fixed employees. However, they are willing to consider introducing IWP in the future.

(23)

Integral Workforce Planning 23 communication. Currently, the in-house employee of Randstad get information about for example sickness of employees at the last moment because the company handles this itself. This communication should be improved.

Another particular aspect is the fact that when an employee is sick, Randstad does not have to provide replacement. The company will internally shift employees from one department to the other. According to the in-house employee this makes no sense because "if this is possible, too many employees are scheduled in the first place." The branch director of Company C made clear that flexibility is an important factor for the company. However, this is not reflected in the way workforce planning is handled.

In an interview with one of the supervisors of Company C the workforce planning process was explained. The demand of the company is very uncertain. This indicates that the amount of orders of next week is unknown. More specifically, the amount of orders of the next day is not even certain. Thus, the planners make their workforce plans based on gut feeling and experience. Therefore, employees are scheduled amply. Generally, the workforce plan is the same every week. The same amount of permanent staff is scheduled every day. This also holds for the temporary employees. The supervisor checks for any extra activities every day. If there are any, he takes this into account in the workforce plan. If the team manager expects a busier period in the next month, he indicates to his staff that they reckon with having to work overtime in that period. These busier periods are considered by analysing the patterns in revenues of previous years.

Until eleven o'clock in the morning orders may arrive. Consequently, the workforce plan can change at the workplace by shifting employees between departments. A planning tool is available in which the amount of employees and hours are shown. It becomes clear if there is any time left. In case there is, employees are shifted to other tasks or flexible employees are sent home. Another option the company takes into account is to work ahead for the next day. By working overtime and working ahead for busier days, the planning process becomes more flexible. This is also achieved by shifting employees between tasks and departments. However, in order to do so the employees need qualifications to perform certain tasks. These qualifications can be obtained by means of (cross) training. These trainings are taken into account in workforce planning as it is a time consuming process. One of the supervisors stated that the time is takes before an employee is fully trained is four to five months. The in-house employee of Randstad explained that this fact causes less flexibility as investing in new temporary workers is expensive. Further, only a couple employees can be trained at the same time. However, within Company C a difference in opinion exist regarding the training time. In an interview with a second supervisor, it became clear that one to two weeks are enough to determine whether someone can stay or has to be replaced.

The team managers of Company C mentioned a factor that is even more important than costs and qualifications: the productivity of the employees. They stated that a balance should be considered between costs and benefits. The team managers stated that costs do not matter if the productivity of an employee is high. This means that a more costly and productive employee will be preferred over an employee that is less costly but also less productive. The cost of scheduling the more productive employee can in the long run result in lower cost and higher benefits.

(24)

Integral Workforce Planning 24 divided more efficient, it will save costs and there will be a better overview of the total workforce. Randstad even sees the possibility to include another location of the company a few kilometres away into the overview to be able to shift employees between the two locations. However, a negative effect could be that instead of planning by name workforce plans will be based on qualifications of the employees. This indicates that preferences of team managers will not be taken into account anymore. In addition, it might cause a change in temporary staff, as not everyone will be willing to work shorter shifts. Further, the branch director explained that for a logistics company the way of dealing with employees is a way to differentiate from its competitors and that the company is concerned that they might lose this uniqueness when delegating the total workforce planning to Randstad. Furthermore, the operations manager mentioned that he would expect Randstad to pay the cost of understaffing in case of IWP. Overall, the team managers seem to be sceptical about the acquisition of the workforce planning process by Randstad. This is mainly caused by not knowing what the process will look like.

Several other factors were found that have not been mentioned explicitly. An overview of all the factors found at Company C can be found in table 6. The quotes in this table are based on the interviews and on questions that were asked during observing the planning process.

Table 6. Factors that are important in IWP based on practical study at Company C

"Every week the same employees are scheduled. Even the flexible staff are scheduled as if they are permanent staff. Every week we [Company C] we schedule the same temporary employees."

"If there are any extra activities, we [Company C] order more employees."

Qualifications of employees "We [Company C] want employees to have as many qualifications as possible so that we can assign them to multiple tasks."

"In order to schedule employees we [Randstad] need to know their qualifications."

Age of employees "Younger employees are less costly. Further, they can also physically perform more tasks. Age is therefore something I take into account when I make workforce plans."

Availability of employees "For us [Randstad] the availability of the employees is important. Otherwise we cannot schedule them in our system."

"We [Company C] find it a requirement for employees to be willing to work overtime."

Productivity of employees "Something that is even more important than qualifications and costs is the productivity of the employees. You cannot only look at the costs, you also have to look at the benefits."

"If someone is better at a certain task than another employee, I will schedule this person."

Teamwork "We want employees that work well in our team. Therefore, we schedule the same employees every day. This also holds for the temporary staff."

"Someone who works well in the team I prefer over someone that is cheaper and does not work well in the team."

Costs

Travel cost "We [Company C] prefer employees that live closer to the company as those employees are cheaper."

Different cost per skill "There are employees that have specific skills and qualifications. Those employees are more expensive and therefore we have to take this into account."

Overtime cost "Overtime is more expensive than normal working hours. We [Company C] therefore try to minimise this overtime as much as possible. However, some overtime can serve as motivation and therefore this is allowed."

Tasks "There are certain tasks that are physically harder to perform. We take this into account."

(25)

Integral Workforce Planning 25 "If the workload at one task is low and at another task the workload is higher, we shift some employees from one task to the other if they have the right qualifications."

Flexibility

Multiple qualifications "Flexibility is obtained by shifting employees between tasks and departments when multiple qualifications are present."

Amount of employees Not applicable. Overtime

Working ahead

"We [Company C] are not flexible in the amount of employees we schedule, but we are flexible because we work overtime and because we work ahead."

(Cross)Training "We prefer employees to have several qualifications. This way we are more flexible. We can therefore cross-train them."

"In order to optimize the workforce planning, more cross-training is needed."

Shifts overlap Not applicable.

Under/overstaffing Not applicable.

"If we have too many employees at work because there are no tasks, we send the flexible employees home, so this is not taken into account while making the workforce plans. It is dealt with at the moment itself."

Preferences

Preferences of employees Not applicable.

"We [Company C] do not take preferences from employees into account. If we would do so we would have 35 lists of constraints. That is impossible. However if an employee always wants to work at the same task we keep this in mind." "We [Randstad] do not consider preferences of employees."

Preferences of unit leaders "There are certain preferences that play a role. I prefer motivated employees over unmotivated employees. I want them to come back to work instead of the unmotivated ones. Also if someone is good at a certain task, I want him or her to work at that task as much as possible."

Time-related "We [Company C] require our employees to be willing to work overtime. As much as needed. Employees who are only willing to work 8 hours a day, we do not want in our company."

Holidays "In case of holidays, we try to schedule all holidays in the given periods. However, if more people want to have a holiday at the same time we discuss who is willing to shift the holiday. We do not hire other flexible employees in order to replace the employees that are on holiday."

Training-time "The time it takes before someone is fully trained and productive is four to five months. This should be considered in workforce planning as it takes a long time to train someone new."

"A training-time of four to five months is way too long. After one or two weeks one can determine if someone can stay or if he or she has to be replaced."

Time horizon "We [Company C] make workforce plans daily."

"The workforce plans for unloading the containers are on a weekly basis."

Demand and forecasts Not applicable.

"Our demand is very uncertain, we do not know in advance how many orders to expect and therefore we do not know how many employees we need. So we [Company C] do not take demand into account as it is only roughly known. We schedule employees based on gut feeling and experience."

Risk "For unloading the containers, reserves are scheduled. One spare employee is scheduled every day."

"If one or more employees are sick, the other employees work overtime. Overtime is a way to cope with this risk."

"As demand is uncertain, we [Company C] plan employees amply."

Communication "We [Company C] discuss everything with our employees. For example things about holidays and expected overtime of more than two hours."

(26)

3.3 Intermediate conclusion

This chapter described the case studies that have been performed. Both the logistics sector and the manufacturing sector were studied. What is striking is that there are several factors that were found in literature that were not used in practice. Further, as shown below, new factors related to workforce planning were found during the case studies that were not found in literature. Between the cases similarities and differences exist.

As for general insights, it can be noted that for all cases that were studied cost was the most important performance measure in workforce planning. However, for the team managers of the companies, this was not their priority and therefore it can be concluded that there are more performance measures that are important. This will be discussed in section 5.1. The performance measure cost will be the basis for the remainder of this research. Further, there are several factors found in literature that were not mentioned in each of the case studies. For example, the shift sequence and overlap were not taken into account as well as the speed of work and breaks. A reason for this could be that in the two cases that already implemented IWP the in-house planners of Randstad oversee the whole process of planning and that they therefore do not take into account relatively small factors such as these. The case that did not implement IWP only had one shift which is the reason why the company does not take, for example, shift overlap into account. Other factors from literature were combined into one factor in practice. Most time-related factors were taken into account by taking into account the law of working hours and were therefore combined into this one new factor.

As for similarities, several factors found in literature were also mentioned in all cases studied. The factor that was mentioned most in all cases was the qualifications of the employees. These qualifications are very important to consider because without qualifications an employee cannot be scheduled for certain tasks that require these qualifications. Other factors that were discussed during the company visits were flexibility and preferences. However, these were further specified. From all cases it became clear that flexibility is an important factor to take into account. However, there are different ways to incorporate this factor in the process of workforce planning. One way of dealing with flexibility is having employees with multiple qualifications. So in this case flexibility is related to training. As mentioned in the introduction, IWP is an interesting concept when it comes to trends as sharing employees. In these trends flexibility is very important and it automatically requires employees to be flexible by having multiple qualifications. Another way in which flexibility is taken into account is being flexible in the amount of employees that are scheduled daily. A last way that was retrieved from the case study is being flexible by working ahead for busier periods. Preferences can be divided in preferences of employees and preferences of the team managers. Preferences of employees were not taken into account in any of the cases. On the other hand, preferences of team managers were taken into account or this could be done if the company wanted to. Besides these factors, there are also factors that were found both in literature and in practice, but which were not mentioned in all three cases. Examples of these factors are the age of the employees and differences in cost per skill.

Integral Workforce Planning