Theory of constraints case study in the make to order environment

Orue, Aitor; Lizarralde, Aitor; Amorrotu, Itxaso; Apaolaza, Unai Article Theory of constraints case study in the make to order environment Journal of Industrial Engineering and Management (JIEM) Provided in Cooperation with: The School of Industrial, Aerospace and Audiovisual Engineering of Terrassa (ESEIAAT), Universitat Politècnica de Catalunya (UPC) Suggested Citation: Orue, Aitor; Lizarralde, Aitor; Amorrotu, Itxaso; Apaolaza, Unai (2021) : Theory of constraints case study in the make to order environment, Journal of Industrial Engineering and Management (JIEM), ISSN 2013-0953, OmniaScience, Barcelona, Vol. 14, Iss. 1, pp. 72-85, https://doi.org/10.3926/jiem.3283 This Version is available at: https://hdl.handle.net/10419/261742 Standard-Nutzungsbedingungen: Die Dokumente auf EconStor dürfen zu eigenen wissenschaftlichen Zwecken und zum Privatgebrauch gespeichert und kopiert werden. Sie dürfen die Dokumente nicht für öffentliche oder kommerzielle Zwecke vervielfältigen, öffentlich ausstellen, öffentlich zugänglich machen, vertreiben oder anderweitig nutzen. Sofern die Verfasser die Dokumente unter Open-Content-Lizenzen (insbesondere CC-Lizenzen) zur Verfügung gestellt haben sollten, gelten abweichend von diesen Nutzungsbedingungen die in der dort genannten Lizenz gewährten Nutzungsrechte. Terms of use: Documents in EconStor may be saved and copied for your personal and scholarly purposes. You are not to copy documents for public or commercial purposes, to exhibit the documents publicly, to make them publicly available on the internet, or to distribute or otherwise use the documents in public. If the documents have been made available under an Open Content Licence (especially Creative Commons Licences), you may exercise further usage rights as specified in the indicated licence. https://creativecommons.org/licenses/by-nc/4.0/ Journal of Industrial Engineering and Management JIEM, 2021 – 14(1): 72-85 – Online ISSN: 2013-0953 – Print ISSN: 2013-8423 https://doi.org/10.3926/jiem.3283 Theory of Constraints Case Study in the Make-to-Order Environment Aitor Orue , Aitor Lizarralde , Itxaso Amorrortu , Unai Apaolaza Mondragon Unibertsitatea (Spain) [email protected], [email protected], [email protected], [email protected] Received: July 2020 Accepted: November 2020 Abstract: Purpose: The theory of constraints (TOC) drum-buffer-rope methodology is appropriate when managing a production plant in complex environments, such as make-to-order (MTO) scenarios. However, some difficulties have been detected in implementing this methodology in such changing environments. This case study analyses a MTO company to identify the key factors that influence the execution of the third step of TOC. It also aims to evaluate in more depth the research started by Lizarralde et al. (2020) and compare the results with the existing literature. Design/methodology/approach: The case study approach is selected as a research methodology because of the need to investigate a current phenomenon in a real environment. Findings: In the case study analysed, the protective capacity of non-bottleneck resources is found to the key factor when subordinating the MTO system to a bottleneck (BN). Furthermore, it coincides with one of the two key factors defined by the literature, namely protective capacity and protective inventory. Originality/value: The three key contributions of this study focus on the MTO environment as follows. The first is about identifying the key factors in subordinating the system to the BN (step 3, TOC) according to the existing literature which have been identified through a systematic literature review. The second focuses on identifying the key factors in subordinating the system to the BN through a case study. Finally, the last contribution compares the results obtained in the case study with those obtained in the literature review. Keywords: MTO, TOC, DBR, five focusing steps; protective capacity To cite this article: Orue, A., Lizarralde, A., Amorrortu, I., & Apaolaza, U. (2021). Theory of Constraints Case Study in the Make-to-Order Environment. Journal of Industrial Engineering and Management, 14(1), 72-85. https://doi.org/10.3926/jiem.3283 1. Introduction In today’s globalised environment, existing market competition has driven many companies to migrate from a make-to-stock (MTS) to a make-to-order (MTO) system, which exhibits continual growth in the number of catalogue references and reduced serial batch (De la Calle, Grus & Álvarez, 2017). In this context, MTO companies lead to non-standard product routing on the shop floor with longer lead times than MTS companies (Stevenson, Hendry & Kingsman, 2005). Accordingly, production planning and control systems (PPCSs) have become key elements in managing material flow in production plants (Jaegler, Jaegler, Burlat, Lamouri & Trentesaux, 2018). -72- Journal of Industrial Engineering and Management – https://doi.org/10.3926/jiem.3283 Several studies have focused on the drum-buffer-rope (DBR) methodology in MTO environments. Theory of constraints (TOC)-DBR has proved to be a valid methodology for responding to the problems of MTO environments (Chakravorty, 2001; Darlington, Francis, Found & Thomas, 2015; Modi, Lowalekar & Bhatta, 2019; Riezebos, Korte & Land, 2003). However, other studies have detected difficulties in implementing TOC-DBR (Atwater & Chakravorty, 2002; Pretorius, 2014; Wu & Yeh, 2006). To overcome these issues, Lizarralde, Apaolaza and Mediavilla (2020) developed a systematic decision-making process to apply DBR in MTO scenarios by proposing a change in the first two steps of TOC methodology. The present research analyses the DBR implementation carried out by Lizarralde et al. (2020); the objective is to identify the key factors of the implementation on the third TOC step and compare the results with the existing literature to advance the proposed systematic decision-making process. This article begins with a literature review in Section 2, where the existing literature on TOC-DBR methodology in MTO companies is analysed with a focus on implementation problems. In Section 3, a systematic review of the literature in the mentioned field is carried out, taking into account that the research will focus on the identification of the key factors when implementing the third step of TOC-DBR. In Section 4, the objectives of the article and the research question (RQ) are defined. Afterwards, in Section 5, the selected research methodology is presented through a case study. The discussion continues in Section 6 with the analysis of the case study, the participation of the researchers and the results obtained. In Section 7, the degree to which the research objectives are achieved is analysed. Finally, the conclusions and future lines of research are presented in Section 8. 2. Literature Review TOC is a management methodology based on systems thinking (Boyd & Gupta, 2004). Its main idea is that every system has at least one constraint that limits its performance. This constraint is taken as the basis for managing and improving the system (Goldratt & Cox, 2004). According to Goldratt and Cox (2004), a constraint or bottleneck (BN) is anything that limits a system from achieving higher performance versus its goal. The five steps of TOC are as follows: (1) identify the system’s BNs, (2) decide how to exploit the BNs, (3) subordinate everything else to the above decision, (4) elevate the system’s BNs and (5) if a BN has been broken in a previous step, return to Step 1. The TOC scheduling mechanism, DBR, is a PPCS in shops with BNs, and it addresses both market and physical constraints (Thürer, Stevenson, Silva & Qu, 2017). To control the whole system, only accuracy in the BN is required, so DBR is a simple approach to carry out (Gupta & Snyder, 2009). Once the restriction is identified, DBR synchronises the production with the customer’s needs through the rope. It connects the work input with the BN (Thürer et al., 2017). The BN programme is scheduled according to the company’s sales portfolio, and the programme criteria are previously defined in the system-exploiting step. Finally, DBR uses drum and shipping buffers (time or amount of time equivalent to work in progress (WIP)) to allow synchronisation and, at the same time, protect the system’s performance from variability with low levels of WIP (Thürer et al., 2017). Generally, nonBN resources are not programmed because each operation is governed by the consumption of the buffer (Goldratt & Cox, 2004). Several studies validated the TOC-DBR methodology for a MTO environment by demonstrating its beneficial effect on workflow performance through reduced lead times and cycle times and increased revenue (Chakravorty, 2001; Darlington et al., 2015; Modi et al., 2019; Riezebos et al., 2003). However, other studies showed the difficulties in implementing the TOC-DBR methodology in this changing environment. Atwater and Chakravorty (2002) defined several issues, such as quantifying the additional capacity of non-BN resources and the appropriate identification of the BN, by considering that real environments represent temporary situations and are not realistic. By contrast, Wu and Yeh (2006) analysed the complexity of implementing the DBR methodology in complex manufacturing environments, focusing on BN re-entry flows and the importance of BN resource sequencing. Finally, Pretorius (2014) identified the shortcomings for decision making that facilitates moving from one step to the next, in addition to the lack of clarity around the ideal location of constraints. Several researchers have worked on a series of alternatives to solve the problems raised in the previous section. Pretorius (2014) proposed a decision map that included the five steps of the TOC methodology; the map defined a -73- Journal of Industrial Engineering and Management – https://doi.org/10.3926/jiem.3283 series of strategic points that help the company make the right decision and allow them to move from one step to the next. By contrast, Lizarralde et al. (2020) provided a strategic perspective for selecting and exploiting BNs. The researchers developed a systematic decision-making process that systematised the first two DBR stages. As shown in Figure 1. Systematic process for steps 1 and 2 of theory of constraints methodology (Lizarralde et al., 2020), the researchers provided a set of four criteria to define how the BN is selected. In addition, three operative sub-steps for deciding how to exploit the BN are included. Figure 1. Systematic process for steps 1 and 2 of theory of constraints methodology (Lizarralde et al., 2020) 3. Systematic Literature Review of TOC Third Step TOC step 3 deals with the resources that are non-constraints, which require them to work in a way that supports the constraint. Non-BN resources have more capacity than constraints do by definition, and thus doing more than the needed work will only produce WIP that the BN is not able to take on. Step 3 only deals with the management of non-BN resources and their level of utilisation, which is determined by the capacity and utilisation of the constraint rather than the potential of the non-constraint resources (Goldratt & Cox, 2004). To deepen the third step of TOC, a systematic review of the literature is carried out. A systematic literature review is conducted to assess and identify research that is relevant to a particular RQ, topic area or phenomenon of interest (Kitchenham, 2004). It is necessary to define a research strategy to carry out an adequate and comprehensive systematic review (Kitchenham, 2004). This literature review uses a strategy based on Kitchenham (2004) and is detailed in Figure 2. Systematic literature review methodology based on Kitchenham (2004). 3.1. Planning the Review This study is motivated by the possibility of deepening the systematic implementation of the TOC-DBR methodology following Lizarralde’s (2020) systematisation of the first two steps. The aim of this literature review is to analyse existing research on the implementation process of the third step of TOC-DBR in MTO environments and, more specifically, the identification of the key factors when subordinating the system to a BN. When planning the literature review, it is necessary to define a protocol that details the methodology used for conducting a specific systematic review. A defined protocol is needed to reduce the effect of researchers’ bias (Kitchenham, 2004). -74- Journal of Industrial Engineering and Management – https://doi.org/10.3926/jiem.3283 Figure 2. Systematic literature review methodology based on Kitchenham (2004) The defined protocol is as follows: Keywords to Carry out the Literature Review The articles focus exclusively on the third step of TOC, which can be defined as the subordination of everything to the BN. Therefore, all the concepts related to this step were entered into the search engine for title, abstract and keywords. All the combinations used are presented below: •‘Drum buffer rope’ OR ‘DBR’ OR ‘theory of constraints’ OR ‘TOC’ AND ‘capacity margin’ •‘Drum buffer rope’ OR ‘DBR’ OR ‘theory of constraints’ OR ‘TOC’ AND ‘subordination’ •‘Drum buffer rope’ OR ‘DBR’ OR ‘theory of constraints’ OR ‘TOC’ AND ‘workload’ •‘Drum buffer rope’ OR ‘DBR’ OR ‘theory of constraints’ AND ‘variability’ •‘Drum buffer rope’ OR ‘DBR’ OR ‘theory of constraints’ OR ‘TOC’ AND ‘step 3’ •‘Drum buffer rope’ OR ‘DBR’ OR ‘theory of constraints’ OR ‘TOC’ AND ‘problematic’ •‘Drum buffer rope’ OR ‘DBR’ OR ‘theory of constraints’ OR ‘TOC’ AND ‘implementation’ AND ‘problem’ •‘Drum buffer rope’ OR ‘DBR’ OR ‘theory of constraints’ OR ‘TOC’ AND ‘implementation’ AND ‘variability’ •‘Drum buffer rope’ OR ‘DBR’ OR ‘theory of constraints’ OR ‘TOC’ AND ‘scheduling problem’ •‘Drum buffer rope’ OR ‘DBR’ OR ‘theory of constraints’ OR ‘TOC’ AND ‘protective inventory’ OR ‘protective capacity’ •‘Bottleneck’ AND ‘protective capacity’ •‘Drum buffer rope’ OR ‘DBR’ OR ‘theory of constraints’ OR ‘TOC’ AND ‘capacity management’ The Sources to Identify Primary Studies Scopus and Web of Science were the databases chosen for the research. They are two of the largest available databases of citations and abstracts of peer-reviewed literature and include all major operations and management publishers that are indexed. Additional potential articles were also identified by searching Google Scholar. -75- Journal of Industrial Engineering and Management – https://doi.org/10.3926/jiem.3283 Select the Exclusion and Inclusion Criteria for the Studies The criteria used to select, analyse and evaluate the articles were as follows: (1) exclusive focus on the third step of TOC; (2) publication in an academic journal or conference; (3) included case studies, both simulated and real; (4) inclusion of no other methodology; and (5) exclusion of articles written for a final bachelor or master's degree. Period of Publication The delimited publication period was from 1990 to 2020. The year 1990 was chosen as the beginning of the publication period because it denotes when the TOC-DBR methodology began to attract the attention of several key authors. The year 2020 was defined as the end date of publications to maximise the number of studies with as many updates as possible. Study Quality Assessment Both quantitative and qualitative documents were considered for this research. The indicators used to evaluate the quality of the selected journals were the Journal Citation Report and SCImago Journal Rank. 3.2. Conduct the Review An unbiased search strategy was defined to find as many primary studies related to the third step of the TOC methodology as possible. Given that few studies make a direct reference to the implementation process of the third step of TOC, the first decision was to select the terms ‘theory of constraints’ or ‘TOC’ and ‘drum buffer rope’ or ‘DBR’. This choice of keywords was intended to ensure that TOC-DBR would be the main theme of the article. To refine the search and focus on the implementation of the third step of TOC, key terminology that relates with the subordination of the system to the BN was added. Figure 3. Literature review selection process illustrates the selection process carried out. Figure 3. Literature review selection process -76- Journal of Industrial Engineering and Management – https://doi.org/10.3926/jiem.3283 Following this process, 41 articles remained, and these were carefully reread again. Figure 4. Frequency of TOC-DBR selected papers, 1990-2020 and Figure 5. TOC-DBR selected articles in journals by publisher show the 41 analysed articles by publication year and journal of publication, respectively. Figure 4. Frequency of TOC-DBR selected papers, 1990-2020 Figure 5. TOC-DBR selected articles in journals by publisher 3.3. Document Review As mentioned in the Section 3, Step 3 deals with the resources that are non-constraints, which require them to work in a way that supports the constraint. This critical constraint should be protected from variation and uncertainty within the system to ensure that the planned throughput is not limited (Patterson, Fredendall & Craighead, 2002). This variability in the system can generate that non-constraint work stations would starve the constraint work station. In other words, the constraint work station can be available to work, but the resource has no work to perform. This condition, called constraint starvation, occurs when the WIP has not reached the constraint work station (Blackstone & Cox, 2002). There are two different methods to reduce starvation in the BN. The first one uses the capacity margin in non-BN resources (protective capacity), and the second one uses the WIP inventory in front of the BN (protective inventory) (Kim, Cox & Mabin, 2010). 3.3.1. Protective Capacity To understand what protective capacity means, it is important to know some basic terms. As revealed in Figure 6. Productive and idle (protective and excess) capacities (Kim et al., 2010), capacity can be classified into two major types, productive and idle. -77- Journal of Industrial Engineering and Management – https://doi.org/10.3926/jiem.3283 Figure 6. Productive and idle (protective and excess) capacities (Kim et al., 2010) Productive capacity is defined as ‘the maximum of the output capabilities of a resource (or series of resources)’ (Pittman & Atwater, 2016: page 146). Idle capacity is defined as ‘the available capacity that exists on non-constraint required to support the constraint. Idle capacity has two components, protective capacity and excess capacity’ (Pittman & Atwater, 2016: page 83). Protective capacity is defined as the ‘resource capacity needed to protect the throughput of the system by ensuring that some capacity above the capacity required to exploit the constraint is available to catch up when disruptions inevitably occur’ (Pittman & Atwater, 2016: page 149). Meanwhile, Blackstone and Cox (2002: page 419) defined protective capacity as ‘the capacity needed at non constraint work stations to restore WIP inventory to the location adjacent to and upstream of the constraint work station to support full utilization of the constraint work station’. From the literature review, it can be concluded that protective capacity plays a key role in the system’s performance. Atwater and Chakravorty (1994) explained that the use of protective capacity is a key factor to achieve faster cycle times while operating at lower inventory levels. Caridi, Cigolini and Farina (2006) defined that protective capacity plays a relevant role in determining productivity. Atwater and Chakravorty (2002) showed that protective capacity at the second most heavily utilised station can improve the system’s performance. Craighead, Patterson and Fredendall (2001) confirmed earlier studies showing that the placement of protective capacity could reduce mean flow time. Lawrence and Buss (1994) reported that higher levels of protective capacity decreased BN shiftiness at all BN utilisation levels. From the studies analysed, there is no doubt about the need for protective capacity in non-BN resources. However, as defined by Tu, Chao, Chang and You (2005), it is very difficult to determine the correct protective capacity. Patti, Watson and Blackstone’s (2008) research article focused on the amount of protective capacity necessary to achieve improvements in production and distribution or the shape of that protective capacity in production systems. Craighead et al. (2001) conducted a systematic investigation of how protective capacity impacts system performance. In the future lines defined by Caridi et al. (2006), protective capacity is described as a viable means for realising faster cycle times and reducing inventory. However, this extra capacity also has a cost connected to it. Hence, there are additional research paths that can be worked on. 3.3.2. Protective Inventory To understand what protective inventory means, we will follow the same logic as that for protective capacity. Inventory has similar implications and definitions as capacity (Blackstone & Cox, 2002). Productive inventory is defined as the amount of WIP inventory (measured in time units) needed to support the constraint until the material can get from the first operation to the constraint (Blackstone & Cox, 2002). -78- Journal of Industrial Engineering and Management – https://doi.org/10.3926/jiem.3283 ‘From a theory of constraints perspective, idle inventory generally consists of protective inventory and excess inventory’ (Pittman & Atwater, 2016: page 83). Protective inventory is defined as ‘the amount of inventory required relative to protective capacity in the system to achieve a specific throughput rate at the constraint’ (Pittman & Atwater, 2016: page 149). Protective inventory has similar concerns as protective capacity. Several studies talk about the need for protective inventory to protect the throughput from variability. Blackstone and Cox (2002) spoke about the amount of protective WIP that is used to reduce the impact of statistical fluctuation at the non-constraint work stations on the constraint work station and system throughput. Betterton and Cox (2009: page 68) indicated that ‘for real lines, protective inventory at any point in time, is inventory upstream of the constraint over and above the deterministic productive number of units. It is WIP inventory that the constraint needs for uninterrupted operation in a nondeterministic system’. There are certain formulas that aim to calculate the productive inventory needed in the system to maximise the throughput. For example, Schragenheim and Ronen (1990) described a process for setting these buffers, which starts by simply dividing the current lead-time allowance of the company in half and using one portion as the constraint buffer and one portion as the shipping buffer. Managers can use this approach to determine initial buffer sizes and make adjustments over time until optimal buffer sizes are identified. However, as indicated by Blackstone and Cox (2002), there is no mathematical approach for defining protective inventory (or protective capacity). Although an adequate level of protection is needed to reduce the impact of statistical fluctuations in the system, the difficulty comes in defining the ‘adequate’ protective inventory (Blackstone & Cox, 2002). As the risk of excess inventory is the increase in WIP and the addition to the production lead time, the minimum levels of protective inventory should therefore be identified (Khalil, Stockton & Fresco, 2008). From the comprehensive literature review analysis, it can clearly be concluded that the levels of protective capacity and protective inventory play key roles against variability. These protection levels will determine the cycle time and throughput of the system. Nevertheless, as Blackstone and Cox (2002) stated in their conclusions, there is no mathematical approach for defining protective inventory and protective capacity. In addition, it is not clear where protective capacity should be placed. Finally, note that a dilemma exists between protective capacity and protective inventory (Figure 7. Dilemma between protective capacity and protective inventory). High levels of protective inventory increase WIP and cycle time, whereas an excess of protective capacity leads to the improper management of the resource. Both situations are not aligned with the TOC philosophy. In consideration of these research areas, the objective of the study is defined in the next section. Figure 7. Dilemma between protective capacity and protective inventory -79-