Analysis of barriers to capacity insights
A single case study
Master thesis, MSc Supply Chain Management
University of Groningen, Faculty of Economics and Business
June 26, 2017
Jochem Sebastiaan Pfeiffer Student number: S2203235 e-mail: j.s.pfeiffer@student.rug.nl
PREFACE
(deleted for confidentiality)
ABSTRACT
Table of Contents
1. INTRODUCTION ... 5
2. LITERATURE REVIEW ... 7
Conceptual model ... 11
3. METHODOLOGY ... 12
Research design: case study ... 12
Case selection... 12
Data collection and analysis... 12
4. CASE DESCRIPTION: PRODUCTION PROCESS ... 15
5. RESULTS ... 16
Capacity and capacity insights ... 16
Data collection and availability ... 18
ERP system and data analysis ... 20
Interdependencies in the production process ... 21
OEE initiatives ... 23
Cultural aspects ... 24
6. DISCUSSION ... 27
Summarizing table and new developed conceptual model ... 32
Research limitations ... 33
7. CONCLUSION ... 34
Suggestions for further research ... 34
REFERENCES ... 35
APPENDICES ... 38
Appendix A: Interview Guide ... 38
Appendix B: Frequency flow diagram ... 41
1. INTRODUCTION
Capacity insights are important in production companies, because the insufficient presentation or lack of information needed for the executions of tasks could cause incorrect decisions in shop floor production planning and control (PPC). These incorrect decisions have a direct impact on the factory efficiencies of the manufacturing organization due to the impact on utilization, deadline adherence and the cost situation (Stowasser, 2006). Capacity insights are based on information which are used to make correct decisions in shop floor planning and control. Capacity insights are related to the level of understanding of capacity availability (by understanding capacity losses) and are related to the level of predictability of the required and available capacity. So, the correct capacity data should be available for the purpose of shop floor planning and control. Many PPC systems and concepts have been developed with the implicit assumption that the information required for key decisions is already in place. However, in practice this is often not the case (Huang, 2017; Stevenson et al., 2005).
The lack or limited availability of this information and thereby the capacity insights mentioned are caused by barriers. While literature acknowledges that in practice capacity data and insights are often not in available (Huang, 2017; Stevenson et al., 2005), no research has been conducted on what barriers are preventing the capacity data and insights to be (limited) available and how these barriers prevent to get capacity data and insights. By knowing which possible barriers to capacity insights there are and how they prevent to get capacity insights, production companies could more easily identify, decrease and erase barriers they encounter. This would result in a higher level of capacity insights which will support correct PPC decision making, which will have a positive influence on the factory efficiencies of the production organization. To both fill this gap in literature and provide aid to production companies lacking capacity insights the following research question is formulated:
What barriers to capacity insights are encountered in production, and how do these barriers prevent to get capacity insights?
2. LITERATURE REVIEW
A literature review is conducted in order to get a better understanding of capacity insights and the possible role of data (information) collection, analysis and availability in shop floor planning and control decisions. There is also a focus on a particular information system which could be used as a mean to gather capacity insights and on manufacturing system interdependencies which increase the systems’ complexity. The OEE tool is also covered to show how it could contribute to a better understanding of capacity including reasons why an organization should not use all OEE elements directly in a capacity analysis.
Production information in planning and control
In a lot of cases production information is (partly) provided by information systems. However, this information is not provided in a timely, structured and practical format at the time the decision-maker needs to make a decision. This forces the decision-maker to browse through different information systems and analyse the information he/she is able to find to make informed decisions. In practice, a decision-maker does not have enough time to obtain sufficient information to make an appropriate decision in a given situation. Resulting in non-optimal shop floor planning and control decisions (Arica et al., 2016)
Tešić et al. (2010) confirm this and state that if production information flows are unavailable for the decision-making processes, poor coordination develops. This will result in a lagging response to changing conditions on the shop floor and ultimately to lost productivity.
Zhong et al. (2015) argue that correct, sufficient and real-time production information could have a significant impact on shop floor planning and control practices. They show that using a RFID-enabled environment using an advanced production planning and scheduling system the average total tardiness could be decreased significantly. This paper shows that smart manufacturing technologies can be used to increase shop floor performance.
Manufacturing system interdependencies
In manufacturing, the concept of dependencies is applicable to a manufacturing process if one considers a process to be a time dependent sequence of elements (Wilson and Platts, 2010). Coordination theory (Crowston, 1997) is viewed as an approach to the study of processes and where coordination is defined as the process of managing dependencies among activities.
Frayret et al. (2004) identifies six types of activity interdependencies in manufacturing systems. These are: pooled, sequential, reciprocal, intensive, task/subtask and simultaneity. They also identify coordination mechanisms which can be used to coordinate each type of dependency. An example of a coordination mechanism used to manage the interdependency of simultaneity is scheduling. More manufacturing system interdependencies could require the need for more coordination mechanisms to coordinate and manage these interdependencies which increases manufacturing complexity.
Manufacturing execution system (MES)
A MES does not actually execute the manufacturing, but rather collects, analyses, integrates and presents the data generated in industrial production so that employees have better insights into processes and can react quickly, leading to predictable manufacturing processes (Naedele et al., 2015). In this way production processes could be optimized (Saenz de Ugarte et al., 2009).
Figure 2.1 illustrates the functions of a MES and its position between the ERP system and the shop floor.
Figure 2.1: MES functionalities and its position (from Saenz de Ugarte et al., 2009)
Overall equipment effectiveness (OEE)
OEE is defined as a measure of total equipment performance, the degree to which the equipment is doing what it is supposed to do (Williamson, 2006). OEE consists out of three components: availability, performance and quality rate of the output. It is used to identify the related losses of equipment for the purpose of improving total asset performance and reliability. It categorizes major losses or reasons for poor performance and therefore provides the basis for setting improvement priorities and beginning of root cause analysis. It can point to hidden capacity in a manufacturing process and lead to balanced flow (Muchiri and Pintelon, 2008). OEE could also be used to track and trace improvements or declines in equipment effectiveness over a period of time (Dal et al., 2000).
Huang et al. (2003) argues that the OEE tool proposed by Nakajima (1998) is only limited to productivity behaviour of individual equipment. Improving OEE is important but is insufficient because no machine is isolated. The manufacturing process is a complex web of interdependencies. This caused modifications and enlargement of the original OEE toll to fit a broader perspective as deemed important in manufacturing systems.
OEE: factors affecting successful data collection
According to Aminuddin et al. (2016) organizations should adapt a data collection method that is not labour intensive in order to reduce operator resistance against data collection. Sohal et al. (2010) confirms by stating that simplicity is required of data capture, storage, display and benchmarking. Today, a mixed data collection method using automated and manual actions is the most used method. User-friendly templates supported by an automatic data collection system provides reliable performance data. It is important to improve data collection and recording methods in order to maintain the credibility of OEE as a performance measure.
In a more general perspective of data collection and performance analysis Huang et al. (2008) argue that the collection of data becomes a bottleneck while the data processing is no longer an issue due to the rapid improvement in computational power. Manual systems of data collection are time consuming, prone to errors, and tedious. As a result, the information does not accurately and promptly reflect the situations and changes. Without up-to-date information, it is impossible to make accurate shop floor decisions, no matter how advanced ERP/MES systems and equipment are. Suggesting that data collection in an accurately automated fashion should prevent it to be the bottleneck of performance analysis.
Capacity data and its relation to OEE measures
A potential by-product of rigorous and routinized measurement of OEE is the opportunity to realize capacity data collection for the purposes of planning and scheduling, shop floor control, understanding how well manufacturing resources are being utilized and strategic considerations (Leachman, 1997; Coelli et al., 2002).
are mix dependent. Idle time losses for instance depend on the production volume, in turn the bottleneck may depend on the product mix. The amount of lost time for setups/changeovers depends on the mix of production, especially if the setup times are quite variable from product to product. Similarly, lost efficiency resulting from less-than-maximum machine loads also depends on production volume and mix. Rework rates for certain products also may be higher than for others. As a result, the OEE performance measures can fluctuate purely as a result of changes in the production schedule.
In general, capacity analysis is performed in terms of unit quantities released into the factory or in terms of unit quantities completed by the factory in various planning periods. Despite the fact that capacity analysis requires a different set of processing times and overall machine efficiency factors than OEE measures the factors can be computed from a common set of component parameters with a common data collection and data maintenance strategy (Leachman, 1997). As a measure better suited for capacity analysis Leachman (1997) proposes the capacity equipment efficiency (CEE) measure, which is able to cope with the mentioned shortcomings of OEE.
Conceptual model
In the literature review conducted several factors are identified which could be possible barriers to capacity insights. Data collection, analysis and availability in the context of capacity insights. The functionality of the ERP system/MES in the context of capacity insights and especially in relation to data collection, analysis and availability. And the production interdependencies, of which the level in related to the complexity and thereby could barrier straightforward capacity insights (e.g. of an individual resource due to dependencies of other resources). Figure 2.2 presents the conceptual model combining the factors which could be possible barriers to capacity insights.
Figure 2.1: conceptual model of factors which could be possible barriers to capacity insights
ERP/MES functionality Data collection Data analysis Data availability Production interdependencies
3. METHODOLOGY
At the moment, there is no or only a limited understanding of the possible barriers to capacity insights. Neither is clear which barriers are encountered, nor is there clarity on the underlying mechanisms. This thesis aims to investigate and identify those barriers and how those barriers prevent to get capacity insights.
Research design: case study
In order to answer the research question a case study is conducted. According to Karlsson (2016) a case study has three strengths. First, relevant theory can be built from the understanding gained through observing actual practice. Second, it enables to gain understanding of the nature and complexity of the researched phenomenon by answering ‘how’ questions. Third, a case study is able to perform exploratory investigations where the variables are unknown and the phenomenon is not fully understood. As a result, the use of a case study for this research will be useful in order to get insights into the barriers to capacity insights. It should be noted that a single case study will probably not identify all possible barriers to capacity insights. But it does give the possibility to perform an in-depth analysis of the underlying mechanisms of the barriers identified in the case company.
Case selection
(deleted for confidentiality)
Data collection and analysis
this research. By means of an internship, observations of everyday business are possible. The observations will be used as explorative input for the interviews by submitting observations to experts (employees). When possible, hard data from the ERP system will be used to support the data collected using the interviews and observations. By using different methods of data collection, there will be a form of methodological triangulation which will support the construct validity of this research.
In order to answer the research question the two mentioned components of capacity insights are combined with the initial findings from the literature review. In table 3.1 the capacity insights components, their possible barriers and the barrier’s components are shown. The barriers and the barrier’s components are based on the conceptual model showed in figure 2.2 of the previous section. The barriers and barrier’s components are covered in the interview guide, which is given in appendix A. By means of the data collection methods mentioned above the barrier’s components are identified. From these components, the barrier in the case company and its effect on capacity insights are identified. The relation between the barriers components, barriers and capacity insights components will explain the mechanisms behind the barriers to capacity insights in the case company.
Capacity insights component
Barrier factor Components of barrier factor
The level of understanding of capacity availability (by understanding capacity losses)
Data collection ERP/MES functionality to collect data. Data collected on capacity losses/availability. Data collection methods.
Data collection on possible losses are: breakdowns, (planned) maintenance, setups, stoppages (starvation, blocking, lack of operator capacity).
Data analysed to gather a higher understanding of capacity availability.
Data analysis methods. OEE initiatives.
Data availability ERP/MES functionality for user friendly easy, central access to data.
Alternative (to ERP/MES) storage of data and its centrality.
Data sharing methods. The level of
control/predictability of the required and available capacity
Production
interdependencies
Product mix.
Different resources (and their capacities). Different possible routings.
Upstream/downstream interdependencies. Supporting role of the ERP/IS to manage interdependencies.
4. CASE DESCRIPTION: PRODUCTION PROCESS
5. RESULTS
In this section, the research results are presented. The results are based on semi-structured and unstructured interviews with employees of the case company, when possible results are supported with additional quantitative data. First the results on the definition and meaning of capacity and capacity insights at the case company are presented. Second the results on data collection, availability and analysis in the context of capacity insights are presented, including the role of the functionality of the ERP system. Third the results on production resources interdependencies are presented. Fourth the role of culture in the context of capacity insights is presented, including the enabling role of the ERP system.
Capacity and capacity insights
According to the interviews the case company defines capacity as the amount of hours’ production per hour/day/week. In which several levels can be distinguished: plant wide, department or individual resource. Departments with multiple resources are theoretically able to do more than 1 hour of production per clock hour because of parallel processing.
In the case company, there are 3 important components used to determine the available capacity and are thereby used in PPC:
Availability of people
In the case company, all resources are operator dependent. So, in order to get a department running operators are required. The operator-machine factor differs per resource, there are resources which can operated simultaneously by one operator. The availability of people is dependent on the scheduling of shifts and holidays, which is done by the team leaders: the responsible person for one or more departments. The schedule of available people is used by the production planner in order to make the production schedule.
Availability of moulds
Availability of departments/resources
The departments and its resources are basically 24/7 available, there are factors which influence the availability of the resources negatively: (planned) maintenance and breakdowns. It is the planners’ responsibility to match the availability of people, the availability of moulds and the availability of resources. They should however take a lower level of components of certain resources into account when planning production in the batch processes: heath connectors, vacuum connectors, space etc.
Next to available capacity there is required capacity. Required capacity is determined by the ERP-planning, priorities, backlog and production decisions (downstream).
Capacity insights
According to the interviews capacity insights have two different purposes. First, it can be used to get a better understanding of capacity availability. Meaning that there is a better understanding of the impact of different losses on capacity availability. Second, it also has to do with the level of predictability there is of the available and required capacity in the context of PPC, which determines the level of control.
One of the production planners state that the long-term gains are: “It will give a better understanding of the overall status of the production process and indicates when expansion or reduction of the gross production capacity will be necessary.” He also states: “Insights could also help use to identify improvement possibilities in order to increase capacity without large investments”.
Consequence of planning with infinite capacity by the ERP system
The ERP system uses MRP logic in order to do a backwards calculation using the predetermined processing times including safety times without considering the available capacity. In reality, this is not feasible because capacity is finite. According to the logistics lead this does not have a big impact because it is possible to prepare and handle peaks in the required capacity, because they are known far in advance. This is confirmed by other employees. However, decisions made by the downstream assembly department could disrupt these preparations. In the section about resource interdependencies this will be further elaborated. As an example, figure 5.1 shows the peaks in the required capacity for the machining department (deleted for confidentiality)
(deleted for confidentiality)
Data collection and availability
The collection and availability of data which the department responsible for PPC can use to get better capacity insights at the composite department of the case company is very limited. Status data of resources and departments is not collected, not real-time and also not afterwards. Status data includes stoppages due to blocking and starvation, when the resource is running, breakdowns, setups and other data on resource status.
the data would also be an issue”. OEE initiatives conducted, including the role of operators, in the case company are discussed later on.
Processing times
Planners use predefined processing times for each product, which are available in the routing documents. One should realize that the predefined processing times do not always match the realized processing times, which are not collected.
Setup times
For the case company, it is hard to determine the amount of setup times and its impact on the effective capacity. Most setup times are sequence dependent. Different products could use the same mould which make a setup/readjustment not or barely necessary. One of the production planners state: “At the moment, there is no matrix available of required setup times between product a and b”. The realized setup times are not logged. Planners now use an estimated setup time based on an average. This makes it impossible to take the sequence dependency into account.
Maintenance and breakdowns
According to the head of the maintenance department the preventive maintenance planning for each department is available. However, according to a production planner this planning is not centrally available. There is no data collected and available on corrective maintenance and breakdowns which could be used in getting insights into capacity.
Scrap
Another factor influencing capacity is scrap, because it is a direct loss of capacity. “By means of a scrum the causes of scrap are investigated and recorded in excel, there is however no link to the impact on capacity related issues”. So, scrap data is collected and available in excel but it is hard to be analysed in the capacity insights context due to the fact that there is some missing functionality in the ERP system, which will be discussed later on.
Minor stoppages
ERP system and data analysis
The ERP system is missing functionality which can be used for easy and effective data analysis. The logistical lead states: “If I want to analyse or answer a certain question I have to manually get hard data out of Baan (the ERP system), which is very labour intensive”. He also states that: “Continuously I have to filter (deleted for confidentiality) data by hand in order to get the information I want”.
At the moment, a lot of data and information is not centrally available. The logistical lead states: “At the moment a lot of people at a lot of different departments at different activities keep track of data using Excel files. Of which I do not get the chance to look at”. According to the logistical lead the result is that a lot of data is not filled into the ERP system. This requires him and others to do additional research and/or asking around for data in order to conduct a quality analysis. It even happens that data must be estimated. In practice, he states that this can result in: “decisions based on what people are telling me instead of based on actual data of the system”.
According to the technical lead planned maintenance could be incorporated into the MRP calculations of the ERP system: “We could use (deleted for confidentiality) for maintenance planning, because it has the functionality”. However, he states that it is not possible with the current hardware: “Because there are so many departments, taking into account planned maintenance in the calculations of Baan would require a lot of computational power. Which we do not have at the moment. This is the reason why we are doing this by hand in Excel”. Because of this the maintenance planning is not centrally available to the production planner, but is it the team leaders’ responsibility to make sure the production planner has knowledge of the planned maintenance. However, the logistical lead states: “Maintenance information is not available or only minimally available, it does not come in a receptacle in which we as logistics can take data from”.
Interdependencies in the production process
As stated before the production process at the composite unit of the case company could be divided into two parts: the front (production line) and the back-end (job shop) of the process. Due to the product-dependent routing in the job shop the amount of relations and interdependencies between the resources increases. Additionally, there is also an inflow from another location and an inflow from the assembly department.
External interdependencies
The logistical lead states that: “There are also routings in which a small assembly activity is conducted between two composite activities in the back of the process”. An assembly planner is responsible for the assembly activity, which means another planner is involved within the planning of the composite planner. This requires an alignment of the two schedules, creating a dependency. Due to this dependency conflicts occur: “There are assembly planners who do not take these interdependencies that much into account, which could result in a misalignment and could cause a backlog for the composite department while I cannot do anything about it”.
Additionally, a lot of activities of the composite department take place upstream in the overall process of the case company in which it produces a lot of ‘parts’ which will be combined into ‘assemblies’ in the assembly department.
The interdependencies mentioned above also apply to (deleted for confidentiality)
Differences between back-end and front
One planner states that: “The job shop environment in the back of the production process makes it nearly impossible to analyse the future status of the departments and orders”. The case company has access to a shop floor control system in which control information can be gathered. This system shows the inventory buffer including the urgency of each order (based on due dates) at each department. In order to get a grip on the future status of the back-end of the process the ERP-planning can be used. However, the production planner states: “The only thing we could hold on to is the planning made by the ERP. However, due to constant deviation from this planning there is no good image of the back of process and on what workload will arrive at the different departments”.
The front of the process uses the same ERP-planning, but is far better to control and predict. One of the production planners state: “In the front of the process there are a lot less interdependencies and I can decide by myself what jobs are released in the process”. “It is a contiguous chain, in which the articles have the same direction”. The consequences of releasing orders in the front of the process are easier to foreseen which makes the future status easier to predict, also because orders are closer to the moment of release.
The front of the process does not have to cope with the inflow from the other location and the assembly department, which gives planners full control over the release of orders.
Due to these differences and harder determination of the future status of the back-end of the process. One of the production planners state: “It is more important to gather insights into capacity in the back of the process where there is an increase of interdependencies, due to this increase it is also more complex to gather capacity insights”.
Hard data
According to the ERP routing data (deleted for confidentiality) The front of the process basically has (deleted for confidentiality)
frequency of the articles. This Pareto chart gives an understanding of the product mix produced at the case company.
(deleted for confidentiality)
Combining the ERP routing and order frequency data, the intensity of the flows between resources in the composite process can be analysed. This results in a frequency flow diagram, which is showed as figure B1 in appendix B. This diagram gives an understanding of the differences between the front and the back-end of the process and the number of significant relationships (interdependencies) there are between processes. (deleted for confidentiality) The big flow is divided over a greater number of smaller flows. This hard data supports the statements of the interviewees.
ERP planning data is used in order to quantify the predictability of the required capacity and the difference in predictability between the front and the back-end of the process, this data is shown in appendix C. Figure 5.3 shows the average standard deviation (and thereby variation) of 3 weeks of the relative change (%) in required capacity between the weeks. To give a good picture there has been chosen to use departments which are in the middle of the front and the back-end of the process. A distinction between the front and back-end departments show that the back-end processes are overall harder to predict due to a higher average standard deviation of the relative change of required capacity.
(deleted for confidentiality)
OEE initiatives
The second initiative was also an operator dependent initiative based on a digital Microsoft Excel system. This system provided easier analysis with increased data registration, more categories to specify the stoppages were available and there were no pre-set time intervals. However, according to an employee who was responsible for the implementation: “The same problems as with the first initiative occurred. The input of data by operators and their team leaders was unreliable and it was unclear how legitimate stoppages were”. In comparison with the first initiative it did provide more insight on setup times and the frequency of certain stoppages.
The third and current initiative is a sensor-based initiative using (deleted for confidentiality) At the moment, it is not used yet in the composite department. The big advantage over the other initiatives is the automated data collection which is only minimally operator/team leader dependent.
Cultural aspects
Different mind-set in the front and back-end of the process
The front and back-end of the process have a different production mind-set. According to the logistical lead there is an overall mind-set focussed on efficiency and productivity and not on service level. In the front of the process the mind-set differs because there is the realization of the relation between the degradation of quality and the time it takes before an order is processed by one of the autoclaves. In the front, there are also minimal possibilities to batch production. Resulting in a mind-set which is focussed in minimizing lead times. In the back-end of the process an individual mind-set focussed on efficiency and productivity is dominant.
in the shop floor, resulting in a decrease in predictability of the required capacity and thereby a decrease in control.
Culture of unauthorised changing of decisions
At the case company, each employee has their own responsibilities. However, in practice employees do not fully hold on to their own responsibilities by also making and changing decisions which are somebody else’s responsibility. The logistical lead states: “There are agreements on who makes which decision, but nobody is sticking to it. If I think your decision is wrong I just change it. This even happens without letting that person know you are changing his or her decision.” This causes employees to make or change decisions for each other, without discussion or authorisation. According to different employees this happens every day for over many years, it has become part of the culture. According to the logistical lead and one of the production planners this behaviour causes a lot of ad-hoc situations. It decreases the predictability of the required and available capacity and thereby a decrease in control. One example is already given in the section about mind-set, wherein unauthorised batching and the changing of the planning sequence occurs. Below there will be two more examples on how this culture has an effect on capacity insights.
Example: Switching of human capacity
The logistical lead and production planner state that there are an extensive amount of changes made in human capacity (availability of people). Team leaders constantly switch operators from one resource to another. They do this because of the individualistic mind-set mentioned earlier: to increase the efficiency and productivity of their own department. In a lot of cases this happens without authorisation or alignment. This decrease the predictability of available capacity and thereby a decrease in control for the logistics department.
Example: unauthorised or unaligned days-off
The culture enabling role of the ERP system
6. DISCUSSION
This research is focussed on identifying and analysing the barriers to capacity insights. The insights gained using a case study are discussed in this section. First, the barriers and their mechanisms encountered in the case company are discussed, which will result in a new developed conceptual model. After the discussion, the research limitations will be discussed.
Barriers to capacity insights
Capacity insights are related to two aspects, the level of understanding of capacity availability (by understanding capacity losses) and the predictability of the required and available capacity. The first does directly give insights in the effective capacity of resources. The second determines the control and possibility to align available and required capacity.
Several (indirect) barriers have been identified in the case company: data collection, analysis and availability, the functionality of the ERP system regarding data collection, analysis and availability, the level of resource interdependencies in the production setting, the culture and the culture enabling role of the ERP system. The collection, analysis and availability of data and the functionality of the ERP system regarding data collection, analysis and availability could have an impact on the level of understanding of the capacity availability. The level of resource interdependencies, the culture and the culture enabling role of the ERP system could have an impact on the level of predictability/control of the required and available capacity. The factors and their mechanisms of the impacts mentioned will be elaborated further down below.
Data collection, analysis and availability barriers
Production planners for example could increase the level of predictability of available capacity if they have access to planned maintenance data. This is allowing them to analyse that data. The increased predictability of when planned maintenance will be happening will increase the control planners have over the available capacity.
The collection of data by operators themselves could cause data to be unreliable due to the lack of commitment, labour-intensive nature of the data collection and human errors. This could cause incomplete or incorrect data. In order to increase the likelihood of the data to be reliable the data collection should be operator independent (or minimally dependent). Unreliable data which is incomplete or incorrect prevents quality analysis from happening, which is a barrier to capacity insights. In this sense, operator dependent data collection causes low quality analysis, which will then become a barrier. Aminuddin et al. (2016) and Sohal et al. (2016) are more reluctant. They argue that simplicity is required in data collection and storage without mentioning the requirement of minimal operator dependent data collection and storage. However, they argue that improvement of data collection and recording methods is necessary in order to improve capacity availability performance measures (as part of OEE). In order to do so and because of the arguments mentioned earlier, minimizing operator dependent data collection and storage would be a logic step. That is why this step is more in line with the general perspective on data collection and performance analysis of Huang et al. (2008), who argue that data collection nowadays is the bottleneck in performance analysis and which should be prevented to do manually due to its time consuming and error sensitive nature.
Missing functionality of the ERP system in data collection, analysis and availability
Without the analyses, there is no high level of capacity insights, which makes the lack of analysis functionality the main barrier. Lastly, the ERP system should provide data availability functionality. Data analysis could not be conducted if data is not centrally (or on the right location) available. Of course, one could think of other systems running next to the ERP which could be used for the collection, availability and analysis of data. For instance, a MES would be suitable for this purpose (Naedele et al., 2015). In the context of capacity insights, a MES should be able to overcome the shortcomings (missing functionality) of the ERP system (Zhong et al., 2012). If such a system is in place, the indirect barrier caused by the missing functionality of the ERP system can be of lower impact. In short, missing ERP functionality could prevent effective and efficient data collection, availability and analysis from happening if there are no alternatives available.
Resource interdependencies barrier
The higher the level of interdependencies between resources and consequently between their capacities the harder it is to get capacity insights. A higher level of interdependencies between resources and their capacity make it harder to determine the future state of those resources due to the increased variables which have an impact on that future state. This will decrease the predictability of required and available capacity, and thereby it also decreases control. A job shop environment is an example of a production setting with a high level of interdependencies. A production line environment is an example of a production setting with a relatively low level of interdependencies. The level of interdependencies is determined by the product mix, the number of different routings, the number of different processes and the number of inflows and outflows from/to external processes in the production setting. So, a high level of interdependencies will be a barrier to capacity insights.
Cultural barrier
Changing decisions causes the predictability of required and available capacity to decrease and thereby decreases the control of the required and available capacity. The unauthorised changing of decisions could cause the available capacity to change on short-term notice by for example: switching of operator capacity, decrease of operator capacity due to unauthorised days-off.
The unauthorised changing of decisions could cause the required capacity to change on short-term notice: changing the production sequence, not allowed batching, decisions made downstream without alignment. Which will cause variation in the production flow, causing it to be unbalanced. Large BOM’s amplify this effect. This phenomenon is in line with the concept of nervousness (Dolgui and Prodhon, 2007; Genin et al., 2007). Which is related to the continual significant adjustments to the production schedule(s). Both argue that freezing the master schedule (for a certain period) within the planning horizon could be a suitable strategy in order to cope with nervousness. However, one should take the mentioned culture into account in order to make such a strategy effective.
Culture enabling role of the ERP system
In the case of the mentioned culture the ERP system could have an enabling role. If the ERP system does not require any form of authorisation when data and/or decisions are changed it could enable and support a culture wherein decisions are changed unauthorised to be executed, which lead to a decreased predictability of the available and required capacity. Even if a ‘freezing’ strategy is implemented it would require the ERP system to ask for authorisation in order to prevent the emergence of nervousness. An ERP system which does require authorisation in order to change data and/or decisions will makes it a lot harder to execute the culture mentioned, and thereby will most likely decrease the effect of the mentioned culture as barrier to capacity insights.
Infinite capacity ERP-planning
They do this by producing part of the orders from a period with relatively high required capacity in a period(s) with relatively low required capacity so the overall required capacity is distributed evenly. Distortions or decisions changed downstream do have a bigger impact on capacity insights in such an environment. In a MTO environment in which the orders are not known (far) in advance the MRP logic planning using infinite capacity could be more problematic, because such a production company does not or only limitedly have the ability to respond by spreading the required capacity in their shop floor planning.
Required attention in literature
Summarizing table and new developed conceptual model
Table 6.1 gives a summary of the discussion.
Factor Barrier Effect on Mechanism
Data collection Indirect Data analysis Missing or low-quality data collection (e.g. due to high operator dependency) causes incorrect or incomplete data which prevents (high quality) analysis
Data analysis Direct Barriers to capacity insights
Labour intensive analysis due to missing or incorrect data (requires data collection by hand) decreases the likelihood of certain analyses to be performed, which cause a barrier to capacity insights. The same goes for low quality analysis because of low quality data collection. Data availability Indirect Data analysis Missing data causes missing or
low-quality data analysis which causes a barrier to capacity insights
ERP system
functionality
Indirect Data collection, analysis and availability
Missing ERP functionality regarding data collection, analysis and availability: causes missing or low-quality data collection, labour intensive or missing analysis and data not available for analysis. Level of resources interdependencies Direct Barriers to capacity insights
A high level of resources interdependencies makes it harder to determine the future state of the shop floor which creates a barrier to capacity insights by decreasing the predictability of required and available capacity.
Culture Direct Barriers to
capacity insights
A culture wherein decisions are changed unauthorised decreases the predictability of required and available capacity and thereby creates a barrier to capacity insights.
Culture enabling role of the ERP system
Indirect The effect of culture on capacity insights
An ERP system which does not require any form of authorisation to change data/decisions in its system enables a culture wherein decisions are changed unauthorised to thrive.
To turn the discussion and table 6.1 into a graphical representation a conceptual model is presented in figure 6.1.
Figure 6.1: conceptual model derived from the discussion
Research limitations
First of all, due to the fact that this research is conducted as a single case study the external validity of the results is limited. In addition to the single case, the (deleted for confidentiality) company does not support the external validity either. Thereby it is expected there will be more (other) barriers to capacity insights or more specific other components in the identified ones in other cases and contexts. An additional limitation is that the concept of capacity insights used in this research is not clearly defined in other literature, which makes it difficult to compare the results of this research with other literature. Another limitation is the limited support of certain results by using hard data. If there was hard data available (e.g. protocols) on the collection, analysis and availability of data in the context of capacity insights, it could have supported the data gathered by the interviews and observations.
Barriers to capacity insights Data collection - Missing - Low quality Data analysis - Labour intensive - Low quality
- Missing data analysis
Data unavailability Missing ERP
functionality
Culture of unauthorised changing of decisions Enabling role ERP
- No authorisation required to change decisions/data of others
7. CONCLUSION
This thesis identified several barriers to capacity insights including their mechanisms. Based on the results, the barriers are put together in a conceptual model (figure 6.1), showing direct and indirect barriers to capacity insights. Especially the barriers to capacity insights caused by a culture of unauthorised changing of decisions and the ERP system enabling to thrive this culture are interesting unexpected results. It is made clear that in the case company there are quite some factors which could cause barriers to capacity insights. Generally, in literature barriers to capacity insights are neglected by not taking them into account or assuming they are not present. However, it is clear that there are a lot of factors which could make determining and predicting the required and available capacity harder to achieve. This requires to reconsider the role of barriers to capacity insights in PPC, both in practice and in research. When the significance of the barriers is determined and recognized, PPC systems and concepts could be improved in order to make them fit better in practice. In the context of smart manufacturing PCC, barriers should also be taken into account. The capacity insights required to successfully run a smart manufacturing system should be reliable and complete. The results show that there might be quite some barriers to tackle in order to do so. However, smart manufacturing also provides new means to tackle part of those barriers.
Suggestions for further research
REFERENCES
Aminuddin, N.A.B. et al., 2016. An analysis of managerial factors affecting the implementation and use of overall equipment effectiveness. International Journal of
Production Research, 5415, pp.20–7543.
Arica, E., Haskins, C. & Strandhagen, J.O., 2016. A framework for production rescheduling in sociotechnical manufacturing environments. Production Planning & Control, pp.1–15.
Coelli, T., Grifell-Tatje, E. and Perelman, S., 2002. Capacity utilisation and profitability: A decomposition of short-run profit efficiency. International Journal of Production Economics, 79(3), pp.261-278.
Crowston, K. (1997), “A coordination theory approach to organizational process design”, Organization Science, Vol. 8 No. 2, pp. 157-75.
Dal, B., Tugwell, P. & Greatbanks, R., 2000. Overall equipment effectiveness as a measure of operational improvement – A practical analysis. International Journal of Operations &
Production Management, 20(7), pp.1488–1502.
Dolgui, A. and Prodhon, C., 2007. Supply planning under uncertainties in MRP environments: A state of the art. Annual Reviews in Control, 31(2), pp.269-279.
Frayret, J.-M., D ’amours, S. & Montreuil, B., 2004. Coordination and control in distributed and agent- based manufacturing systems. Production Planning & Control, 15(1), pp.42–54.
Genin, P., Thomas, A. and Lamouri, S., 2007. How to manage robust tactical planning with an APS (Advanced Planning Systems). Journal of Intelligent Manufacturing, 18(2), pp.209-221.
Huang, Y., 2017. Information architecture for effective Workload Control: an insight from a successful implementation. Production Planning & Control, 28(5), pp.351-366.
Huang, G.Q., Zhang, Y.F. & Jiang, P.Y., 2008. RFID-based wireless manufacturing for real time management of job shop WIP inventories. International Journal of Advanced
Manufacturing Technology, 36(7–8), pp.752–764.
Karlsson, C. ed., 2016. Research Methods for Operations Management. Routledge.
Leachman, R.C., 1997. Closed-loop measurement of equipment efficiency and equipment capacity. IEEE Transactions on Semiconductor Manufacturing, 10(1), pp.84–97.
Muchiri, P. & Pintelon, L., 2008. Performance measurement using overall equipment effectiveness (OEE): literature review and practical application discussion. International
Journal of Production Research, 46(13), pp.3517–3535.
Naedele, M., Chen, H.M., Kazman, R., Cai, Y., Xiao, L. and Silva, C.V., 2015. Manufacturing execution systems: A vision for managing software development. Journal of
systems and Software, 101, pp.59-68.
Nakajima, S., 1988. Introduction to TPM: Total Productive Maintenance. (Translation). Productivity Press, Inc., 1988, p.129.
Saenz de Ugarte, B., Artiba, A. & Pellerin, R., 2009. Manufacturing execution system – a literature review. Production Planning & Control, 20(6), pp.525–539.
Sohal, A., Olhager, J., O’Neill, P. and Prajogo, D., 2010. Implementation of OEE–issues and challenges. Competitive and Sustainable Manufacturing Products and Services, pp.1-8.
Stevenson, M., Hendry, L.C. and Kingsman, B.G., 2005. A review of production planning and control: the applicability of key concepts to the make-to-order industry. International journal of production research, 43(5), pp.869-898.
Stowasser, S. (2006). Computer-supported visualisation of distributed production area information. Production Planning and Control, 17(2), 202–213.
Victor, B. and Blackburn, R.S. (1987), “Interdependence: an alternative conceptualization”, Academy of Management Review, Vol. 12 No. 3, pp. 486-98.
Williamson, R.M., 2006. Using overall equipment effectiveness: the metric and the measures. Strategic Work System, Inc, pp.1-6.
Wilson, S. and Platts, K., 2010. How do companies achieve mix flexibility?. International
Journal of Operations & Production Management, 30(9), pp.978-1003.
Zhong, R.Y., Dai, Q.Y., Qu, T., Hu, G.J. and Huang, G.Q., 2013. RFID-enabled real-time manufacturing execution system for mass-customization production. Robotics and
Computer-Integrated Manufacturing, 29(2), pp.283-292.
APPENDICES
Appendix A: Interview Guide
This interview guide describes the characteristics and questions used in the semi-structured interviews conducted for the purpose of research. The research is conducted for a master thesis. Results from the interviews will be used to answer the research question of the master thesis: What barriers to capacity insights are encountered in production, and how do these
barriers prevent to get capacity insights?
Capacity insights component
Barrier factor Components of barrier factor
The level of understanding of capacity availability (by understanding capacity losses)
Data collection ERP/MES functionality to collect data. Data collected on capacity losses/availability. Data collection methods.
Data collection on possible losses are: breakdowns, (planned) maintenance, setups, stoppages (starvation, blocking, lack of operator capacity).
Data analysis ERP/MES functionality to analyse data
Data analysed to gather a higher understanding of capacity availability.
Data analysis methods.
Data availability ERP/MES functionality for user friendly easy, central access to data.
Alternative (to ERP/MES) storage of data and its centrality.
Data sharing methods. OEE initiatives. The level of
control/predictability of the required and available capacity
Production
interdependencies
Product mix.
Different resources (and their capacities). Different possible routings.
Upstream/downstream interdependencies.
Supporting role of the ERP/IS to manage interdependencies.
Table A1: factors used for the initial collection of data
Below a general interview framework is given. For each interviewee, the framework is adjusted in such a way the questions emphasis on the function of the interviewee. In most cases the function of the interviewee or his responsibilities are known in advance by asking around.
Appendix B: Frequency flow diagram
