Lean waste prioritisation and reduction in the apparel industry: application of waste assessment model and value stream mapping

Abstract

Manufacturing industries implement lean production tools to improve their production efficiencies. Improving production efficiency is vital in the apparel industry because product cycles and lead times are getting shorter and shorter. The purpose of this study was to improve the performance of a shirt manufacturing company through the implementation of value stream mapping (VSM) along with some other lean tools. Data were collected through stop-watch-based time studies and structured questionnaires. On the one hand, the current state value stream map was created and its performance was examined. On the other hand, the ranks of the seven lean wastes were determined, and process activity mapping was used to analyse the wastes furthermore. Then, layout improvement and 5S were implemented to reduce the main wastes, and the future state value stream map was developed. Analysis of the wastes resulted in motion, inventory and overproduction as the main wastes, while materials and machines were found to be the main sources of lean wastes. The main results of lean implementation showed 74.30%, 30.95%, 71.1% and 27.02% reductions in work-in-process inventory, value-added time, lead time and standard minute value, respectively. The findings of this study support the applicability of VSM in the apparel industry.

Keywords:

• Waste assessment
• prioritisation
• reduction
• apparel industry
• value stream mapping
• VSM
• lean tools

Reviewing Editor:

• Zude Zhou, Wuhan University of Technology, China

Subjects:

• Textile Manufacturing
• Engineering Management
• Production Engineering
• Industrial Engineering & Manufacturing

1. Introduction

Lean production principles, initiated and first implemented by the Toyota production system (Ohno, 1988), have been widely applied in many manufacturing (Bashar et al., 2021; Zahraee, 2016) and service (Kanakana, 2013) industries in different parts of the world. Lean production is characterised by performing more with less use of resources (Womack & Jones, 2003) which can be achieved by implementing lean production tools. The most commonly implemented lean tools include 5S (Jaca et al., 2014), just-in-time (Cuellar-Valer et al., 2020), Kanban (Carvalho et al., 2019), Kaizen (Georgise & Mindaye, 2020) and value stream mapping (B. Singh & Sharma, 2009) among several others (R. Kumar & Kumar, 2012). These tools help in waste reduction (Suhardi et al., 2020), productivity improvement (Goshime et al., 2019) and improvement of the health and safety of workers (Hamja et al., 2019).

While the Toyota production system (Ohno, 1988) and, later on, the western world (Goshime et al., 2019) enjoyed the benefits of implementing lean tools, most developing countries still rely on traditional production systems. Studies show that most of the authors of papers on lean production are from developed countries (Bashar et al., 2021; Danese et al., 2018). Countries with emerging economies such as Ethiopia need to explore and learn ways of contextual implementation of lean tools to replicate the success of developed countries in lean production (Danese et al., 2018; Panizzolo et al., 2012).

The manufacturing industry in Ethiopia has sluggish growth and poor capacity utilisation and most employees, operators as well as managers, are unaware of most of the continuous improvement tools (Tesfaye, 2019). Although Ethiopia has shown some successes in creating policy support for manufacturing industries, especially for the textile and apparel sector, there are still some challenges in the apparel sector including the focus on cut, make and trim (CMT) production, long lead times, low production and product flexibility and limited operator skills (Staritz et al., 2019). To break out of such problems, there are some attempts of lean implementation such as Kaizen in industries in the southern region (Georgise & Mindaye, 2020), work standardisation and line balancing in a garment manufacturing company (Mulugeta, 2021) and value stream mapping (VSM) in footwear (Reda & Dvivedi, 2021) and machine building (Gebeyehu et al., 2022) industries. While these initiations are so good, the limitation in the number of studies and their scope in terms of the lean tools implemented and the product types investigated indicate that much work is yet to be done with regard to saving resources and minimising waste.

This study attempts to contribute to this gap by applying a combination of diagnostic (VSM), ranking (waste assessment model (WAM)) and improvement (layout improvement and 5S) lean tools (Jana & Tiwari, 2021b) to identify and reduce the most prevalent lean wastes in a shirt manufacturing company. The VSM technique is often used with WAM and value stream analysis tools (VALSATs) (Kosasih et al., 2019), but its application in the apparel industry has been overlooked by prior studies (Lugert et al., 2018; Romero & Arce, 2017). Moreover, we used a bundle of lean tools following the success reports of using groups of lean tools than individual tools in the literature (Hodge et al., 2011).

The waste assessment model proposed by Rawabdeh (2005) was applied to identify the prevalent lean wastes and opportunities for waste elimination. Moreover, the result of the lean waste relationship and assessment was used to select an appropriate VALSAT (Hines & Rich, 1997). The VSM method was used to visualise the flow of information, value-adding and non-value-adding activities of the current and desired future state. The transformation from the current to the future state was possible by implementing layout improvement (B. S. Kumar & Sampath, 2012) and 5S (Jaca et al., 2014) which were selected based on the examination of the current state value stream map and the main wastes obtained from the WAM and applied according to the guidelines in Rother and Shook (2003) to reduce waste and shorten lead time. Waste reduction through minimal investment in implementing lean tools has implications for improving productivity, enhancing customer satisfaction and rising the profitability of apparel manufacturing companies. The method can also be extrapolated to related industries to improve resource utilisation through waste reduction.

2. Literature review

Lean production has been adopted by various industries in different parts of the world since Womack et al. (2007) made the term lean popular. It has influenced our daily lives in many aspects besides improving organisations’ structure and operations. Lean tools improve the efficiency of organisations by eliminating the seven production wastes: overproduction, waiting, unnecessary transport, over-processing, inventories, motion and defective parts and products (see Table 1). As wastes are non-value adding to customers, waste minimisation increases the efficiency of organisations and the satisfaction of their customers. To minimise waste, understanding the seven production wastes in detail (Womack & Jones, 2003), prioritising the wastes (Rawabdeh, 2005) and selecting appropriate lean tools (Manickam & Rathinasamy, 2022) are very important.

Table 1. Description of the types of wastes in manufacturing processes (Obeidat et al., 2012; Ohno, 1988; Rahmanasari et al., 2021; Wu, 2003).

Type of waste	Description
Overproduction	It is manufacturing parts and products that are beyond the demanded amount by the subsequent process and the end customer, respectively. It results in an increased number of employees creating more wastes and quality problems.
Waiting	It is the time waste that arises from employees looking for upstream processes or activities to finish their work or until machines perform some activities or get ready
Transportation	It is unnecessary transport of materials between facilities because of improper layout
Over-processing	It is the waste created by poor tools, bad product design, unclear customer requirement, and defective products demanding longer time and effort than normal
Motion	Results from unnecessary movement of employees looking for parts, tools, and help in awkward postures resulting stress and related compensation expenses
Unnecessary inventory	Excessive inventory created by overproduction ties up money and creates cluttered workstation
Defects	Created by not producing the right products the first time and looking for corrective actions

2.1. Lean approaches in manufacturing industry

Suárez‐Barraza and Ramis‐Pujol (2012) explored the contextual implementation of 5S in multinational companies in Mexico and found that the strategic link of the 5S effort, implementation and institutionalisation and self-discipline are the drivers of successful implementation and its sustainability. J. Singh et al. (2018) reported monetary savings as a result of lean manufacturing techniques in the manufacturing industry of northern India. Studies also showed a link between the implementation of certain lean tools such as 5S and poka-yoke and the improvement of the safety and health of workers (C. Singh et al., 2021). Danese et al. (2018) reviewed recent studies on lean production and found that most of the studies are from the USA, UK and India, and they recommended the investigation of lean implementation in less explored countries. Because, studies show that failure to implement lean bundles is likely to put organisations at a performance disadvantage compared to organisations that implement lean systems regardless of the organisational context (Shah & Ward, 2003).

The continual strive for improving the productivity of manufacturing industries has resulted in the development of several lean tools (Goshime et al., 2019; Jana & Tiwari, 2021b). Zahraee (2016) found that 5S, Kaizen, setup time reduction, continuous flow, cellular manufacturing, product design simplicity, equipment layout and error-proof equipment are crucial to lean production. According to Jana and Tiwari (2021b), the nature of lean tools can be broadly classified into diagnostic, improvement and monitoring tools. One of the powerful diagnostic lean tools useful for identifying wastes is value stream mapping which visualises the flow of information and materials in a diagram clearly (Jana & Tiwari, 2021b; Rother & Shook, 2003). Value stream refers to the activities required to design, manufacture and provide a specific product from raw materials to the hands of the customer. VSM is the process of identifying and presenting all specific activities occurring along a value stream for a product or product family (Womack & Jones, 2003). The activities in a value stream include value-adding (VA), necessary non-value-adding (NNVA) and non-value-adding (NVA). To visualise the non-value-adding activities and pave a way for their elimination, VSM has been used along with WAM and VALSATs (Kosasih et al., 2019), simulation (Abdulmalek & Rajgopal, 2007; Helleno et al., 2015; Zahraee et al., 2020, 2021), sustainability (Mishra et al., 2020; Phuong & Guidat, 2018) and other lean tools (Obeidat et al., 2012).

The waste relationship matrix and the rank of lean wastes from the analysis of WAM help to identify the main wastes (Rawabdeh, 2005). Kosasih et al. (2019) used the WAM to prioritise the seven lean wastes and the weights of the seven wastes, which determines their ranks, to select an appropriate VALSAT to analyse the seven wastes. The alternative VALSATs include process activity mapping, supply chain response matrix, production variety funnel, quality filter mapping, demand application mapping, decision point analysis and physical structure (Hines & Rich, 1997). Suhardi et al. (2020) applied a similar procedure to implement VSM in a bra-manufacturing garment company. After identifying and ranking the wastes, improvement and monitoring lean tools and lean metrics are applied to bring changes and sustain it. Table 2 classifies some lean tools applied in manufacturing industries for various purposes.

Table 2. Purpose and category of some lean tools implemented in manufacturing industries.

Category	Purpose	Lean tools	Sources
Diagnostic tools	To identify wastes and areas of improvement	VSM, affinity diagram, Gemba, failure mode and effect analysis	Abdulmalek and Rajgopal (2007), Helleno et al. (2015), Jana and Tiwari (2021b), Rother and Shook (2003), Zahraee et al. (2021), Zahraee et al. (2020)
Prioritisation tools	To rank the wastes and identify the priority areas of improvement	Waste assessment model, VALSATs, Pareto chart	Hines and Rich (1997), Jana and Tiwari (2021b), Kosasih et al. (2019), Rawabdeh (2005)
Improvement tools	To improve the workplace and make visualisation easy	Kaizen, 5S, single minute exchange of die, just in time, line balancing, layout adjustment	Barzola-Cisneros et al. (2021), Ferdousi and Ahmed (2009), Hodge et al. (2011), Jana and Tiwari (2021b), Karunarathne and Dhananjana (2022), Makhija et al. (2021), Ratnayake and Dinosha (2018), Rother and Shook (2003), Silva (2012), Silva et al. (2011)
Monitoring tools	To follow up and sustain the improvement	First in first out, Kanban, dashboard, Gantt chart	Jana and Tiwari (2021a), S. Prasad & Panghal (2022), Reda and Dvivedi (2021), Yu and Kuan (2011)
Performance metrics	To measure the improvements	Lead time, customer takt time, cycle time, throughput time	Behnam et al. (2018), Jana and Tiwari (2021b), Reda and Dvivedi (2021), Rother and Shook (2003)

2.1.1. Lean implementation in the apparel industry

The apparel industry is a labour-intensive industry characterised by many different clothe making processes including fabric inspection, pattern making, marker making, fabric spreading, cutting, sewing/assembling, finishing and packaging, which require considerable human attention, low-level technology and little automation (Hamja et al., 2019; Taplin et al., 2003; Wickramasinghe & Perera, 2016). In this industry, though the demand is ever-increasing, the competition is fierce to win customers in which cutting costs is one of the strategies. Due to this, offshore production of apparel in developing countries with labour cost advantages is widely practiced (Scott, 2006; Taplin et al., 2003). However, low labour skills, high waste, labour turnover and eventual firm turnover affect the production of apparel in developing countries (Panizzolo et al., 2012; Taplin et al., 2003). Hence, the industry is challenged to create a new labour-intensive production model (Wickramasinghe & Wickramasinghe, 2017) which cuts costs (Wickramasinghe & Perera, 2016) and shortens lead time (Ünal & Bilget, 2021) to survive in the international market. The lean production system is then needed to realise the new labour-intensive production model.

Although the labour intensiveness of these industries along with their nature which demands less automation makes them more favourable for the implementation of lean production, the practice of lean ways in the apparel industry is yet limited (Bashar et al., 2021; Jana & Tiwari, 2021a). For example, out of 178 garment factories surveyed in Bangladesh, Bashar and Hasin (2018) found that the average awareness level of lean tools was about 72%, and only 52% of the garment factories implemented lean tools to some extent with only 14% of them agreed that their implementation level was high. However, there are some studies with success stories of lean production in the apparel industry (Wickramasinghe & Perera, 2016; Wickramasinghe & Wickramasinghe, 2017). Implementation of lean production in Bangladesh’s apparel industry resulted in waste elimination (Bashar & Hasin, 2016), successful changes in product design, production processes and supplier networks (Ferdousi & Ahmed, 2009) and impacted the operational and business performances (Bashar et al., 2021). A survey of export-oriented Sri Lankan textile and apparel firms revealed that total productive maintenance (TPM) practices improved cost-effectiveness, product quality, delivery time and flexibility in volume of production (Wickramasinghe & Perera, 2016).

The textile and apparel industry in Ethiopia is one of the priority sectors. The emphasis given by the government is resulting in an increasing number of local and foreign direct investments in the sector. However, low productivity, high production cost, poor quality and firm turnover are some of the problems which need improvement (Staritz et al., 2019; Weldesilassie et al., 2017). Too much motion of workers and transport of workpieces, high inventory and unbalanced work assignments are also found as the main problems in garment manufacturing case companies in Ethiopia (Mulugeta, 2021; S. Prasad & Panghal, 2022). While lean tools such as 5S, VSM, Kaizen, visual management and TPM are commonly implemented in the apparel industry in several countries (Hamja et al., 2019), such studies are limited in the case of Africa, in general, and Ethiopia, in particular (Goshime et al., 2019).

2.2. Application of value stream mapping in manufacturing industry

VSM is a versatile lean tool that is widely applied in manufacturing industries to improve productivity. Through the application of VSM in a vehicle assembly line, Zahraee et al. (2014) improved production lead time and value-added time by nearly 80 and 12%, respectively. Rohani and Zahraee (2015) implemented VSM along with 5S, Kanban, and Kaizen in a colour industry and achieved 29.40 and 45.60% reductions in lead time and value-added time, respectively. Romero and Arce (2017) reviewed the application of VSM in manufacturing industries. They found that more than 44% of the works emphasised on the application of other lean tools along with VSM and used the reduction of lead time as a key performance indicator. Heravi and Firoozi (2017) implemented VSM in the production phase of prefabricated steel frames and reported 34 and 16% reductions in production lead time and costs, respectively. Lugert et al. (2018) assessed the application of VSM in manufacturing industries and noted that VSM is becoming the centre of day-to-day business processes. Deshkar et al. (2018) implemented lean production in a plastic bag manufacturing unit using VSM and achieved a 74.85% increment in percentage of value-added time and a takt time reduction of 20.6 min. Nihlah and Immawan (2018) applied VSM and VALSAT in small and medium manufacturing enterprises and brought 80 min reduction in lead time and a 2.09% reduction in NVA. Kosasih et al. (2019) used VSM and Kanban systems in a chemical company and reported a 67.25% improvement in process cycle efficiency and a 6.74% reduction in NVA. Masuti and Dabade (2019) implemented Kaizen and VSM in an excavator manufacturing company and achieved reductions of 156 min in value-added activities and 430.3 min in NVAs. Dadashnejad and Valmohammadi (2019) investigated the effect of VSM on overall equipment effectiveness and noted that VSM can enhance the rate of machine availability, performance and product quality.

Recently, Zahraee et al. (2020) reported 37.14, 29.22 and 23.20% reductions in production lead time, NVA and takt time, respectively, by applying VSM and simulation in a heater industry. Yadrifil et al. (2020) reported 20.75 and 32.63% reductions in cycle time and lead time, respectively, through the application of VSM and WAM in laminating process of wooden doors. Jasti et al. (2020) implemented VSM in an auto-ancillary industry and achieved a 1-min and 408 min reduction in takt time and lead time, respectively, and an increment of 10% in process ratio. J. Singh et al. (2020) applied VSM and layout improvement in a manufacturing unit and reported 14.88, 14.71 and 37.97% reductions in lead time, processing time and wastage of material movement, respectively. Mishra et al. (2020) developed a sustainable VSM for a bonnet-manufacturing industry and achieved 30 and 83.70% reductions in cycle time and carbon footprint, respectively. Reda and Dvivedi (2021) applied VSM in footwear manufacturing and brought 56.30 and 69.70% reductions in cycle time and lead time, respectively. Zahraee et al. (2021) applied VSM with simulation in a construction industry and reported 36.36, 41.10 and 32.60% reductions in production lead time, NVA and takt time, respectively. Sangwa and Sangwan (2023) improved the line efficiency of an assembly line in an automotive industry by 5.1% and reduced the cycle time by 6.3% through the application of VSM along with Kaizen, Gemba walk and 5 whys.

2.2.1. Literature gap

Among manufacturing industries, the current study considers a garment/apparel manufacturing company. Lean production practice is limited in the apparel industry (Bashar et al., 2021; Jana & Tiwari, 2021a). Studies show that the shortcomings of the traditional management system in the apparel industry of Bangladesh and the cotton spinning industry of Ethiopia have led to declined demand and the obsolescence of production technology and techniques in the textilesub-sector in Ghana and the Italian luxury fashion industry has led to the loss of productivity (George et al., 2022). Compared to other industries, the share of application of VSM in the textile and apparel industry is low (Romero & Arce, 2017). Although there are some attempts (Akçagün et al., 2012; Hodge et al., 2011; Marudhamuthu & Pillai, 2011; Suhardi et al., 2020), VSM is not widely applied in the production process of apparels. Table 3 presents a review of the application of VSM in the textile and apparel industries. The review shows that VSM was not applied to production processes in the apparel industry with many operations such as shirt manufacturing which is affected by relatively higher NVA activities. Moreover, no related study was conducted in the textile and apparel industry in Ethiopia.

Table 3. Review of applications of VSM in the textile and apparel industry.

Authors	Country	Method	Lean tools	Section	Findings
Hodge et al. (2011)	USA	Qualitative interview, observation, and case studies	VSM (two case studies) 5S (three case studies)	Spinning, weaving, and knitting	A five-stage hierarchical approach to lean implementation was proposed VSM can be used for all levels of the lean implementation
Marudhamuthu and Pillai (2011)	India	Case study	VSM and single minute exchange of die (SMED)	Manufacturing (cutting and sewing)	The time to change the setup from one style to another was reduced from 28 min to around 8 min 58 min and 48 s of non-value-adding time was saved per shift
Akçagün et al. (2012)	Turkey	Case study	VSM	From raw materials to finished goods warehouse	The time to accomplish all processes of outwear products was reduced from 27–33 weeks to 9–14 weeks
B. S. Kumar & Sampath (2012)	India	Case study	VSM, cellular layout	Sewing	A reduction in WIP of a T-shirt manufacturing increased the profit by 20–25 % from the existing level
Obeidat et al. (2012)	Jordan	Case study	VSM, line balancing, layout design	Sewing	Results showed a reduction of 96% in production wastes and 43% in production lead time
Silva (2012)	Sri Lanka	Case study	VSM, 5S, one-piece flow, cellular manufacturing	Processes from supplier to customer	VSM can be applied to the apparel industry as the initial step of waste identification It helped the managers to visualise the different types of waste generated in the organisation and future possibilities of eliminating or reducing them
Sultana and Islam (2013)	Bangladesh	Quantitative (13 garments industries)	VSM	From spreading to sewing and finishing	Value addition increased from 15.71 % to 69.55 % on average The benefits of VSM arise from the interaction of the people making the maps with the process and observing on the shop floor
Yildiz and Güner (2013)	Turkey	Case study	VSM	Processes from supplier to customer	A 9.782 days production lead time of an order for jeans products was reduced to 3.4 days
S. Kumar and Thavaraj (2015)	India	Case study	VSM, cellular layout, single-piece flow, and Kaizen	Sewing section	Improved process ratio from 3.73 to 12.71% About 36.25% reduction in standard minutes value (SMV) Grouped operations of knitted men’s wear from 16 to 8 Reduced WIP inventory by 72.97%
Behnam et al. (2018)	–	Case study	VSM, analytical hierarchy process	From fabric cutting to garment packing	Mudas in the production line of clothing factory were identified and prioritised according to seven categories of Mudas using an analytical hierarchy process Defects such as failure in enzyme process and cutting wastes, and waiting wastes such as enzyme shop waiting, delay in receiving fabric from supplier, and delay in fabric cutting were the top five wastes Suggested a future state value stream map to improve the priority wastes
Phuong and Guidat (2018)	–	Case study	Sustainable VSM, radio-frequency identification (RFID)	From cutting to quality control and packing	RFID gives managers the chance for a distant monitoring Due to the RFID, the SSVM can run in real-time and the figures can be tracked at any time Presents a methodology for SSVM and implementation of industry 4.0
Tapia-Leon et al. (2019)	Peru		VSM, line balancing, 5S, Kaizen, layout improvement		The lead time decreased by three days The cycle time decreased by 9 min The downtimes decreased by 25% Order fulfilment increased by 15%
Carvalho et al. (2019)	Brazil	Exploratory case study	VSM, Kanban	From fabric spreading to sewing	The main wastes were found to be inventory and waiting The efficiency of the cycle (dividing VA time by lead time) has been improved from 27.67% to 40.1% after the implementation of Kanban
Suhardi et al. (2020)	Indonesia	Waste waiting questionnaire Time study	VSM, WAM	Cutting, sewing, packing	Motion with a 21.15% contribution was found to be the main waste After the implementation of the future VSM, the total time for the selected Bra production process was 3 min faster than before the improvement The average value of the selected Bra model line production efficiency increased by 6.17%
Cuellar-Valer et al. (2020)	Peru	Simulation	VSM, SMED, TQM, JIT	Cutting and sewing	The VSM and SMED diagnostic tools were first used to standardise the analysed process (cut and sew) The joint implementation of the lean tools improved significantly the service level, quality, and non-compliant products
M. M. Prasad et al. (2020)	India	Case study	VSM, 5S, Kanban, Kaizen, Poka-yoke, Visual controls	Fabric manufacturing	Reduced warp and weft breaks Improved organisation of working environment Motivated workers Improved efficiency of weaving machines
Barzola-Cisneros et al. (2021)	Peru	Case study	SIPOC, VSM, 5S	SME: T-shirt manufacturing	Improved productivity by 5% in the first two months 35% improvement in the level of compliance: timely compliance Satisfaction of final customer increased by 5%
Hartini et al. (2021)	Indonesia	Case study	Sustainable – VSM, 5S	Natural dyeing batik production	Reduced setup time and transportation time Improved working environment Reduced material consumption and increased recovered waste Reduction in production costs
Akter et al. (2022)	Bangladesh	Multiple-case study	VSM, material stream mapping	Textile and apparel manufacturing	Identifies, categorises, stream maps and determines the value addition of materials waste Describes the potentials for empirical research extensions and materials waste management interventions in the textile and apparel industry
Hussain and Figueiredo (2023)	Pakistan	Case study	VSM	Preparatory stage of fabric manufacturing	Improvements in terms of available capacity utilisation, balancing work flow in the preparatory stages, visibility of the process by measuring its capability and flexibility for the planning function
Santos et al. (2023)	–	Case study	VSM, Gemba, 5S, Kaizen, Kanban, Poka-Yoke	Upholstery industry of vehicle seats	Process waste minimised by about 47% Process lead time decreased by about 26% Production increased by 33%

3. Methodology

3.1. Research approach

This paper is based on a case study of lean implementation in a selected garment manufacturing company. The case study was conducted by observation of the actual production floor through time studies and surveys of lean waste relationship and assessment. The research framework is presented in Figure 1.

Figure 1. Research framework.

3.2. Sampling and data collection

Primary data were collected in a stepwise manner. Using a stopwatch, time studies were conducted twice to create the current and future state value stream maps. In each time study, 30 measurements of a complete cycle (45 operations) were recorded (Suhardi et al., 2020). Furthermore, a waste assessment was conducted to identify the main types of lean wastes and select an appropriate VALSAT. The waste relationship and waste assessment questionnaires were completed by 30 employees from cutting, sewing, finishing, production planning and control, warehouse, industrial engineering and management of the case company who have sufficient know-how about the types of lean wastes. A small sample size was used because the questions are many and their answers require deep understanding in which only a few employees can have this expertise.

3.3. Data collection instruments

The time studies were conducted using a stopwatch and a recording format for 45 operations. A waste relationship questionnaire which consists of six structured questions for each relationship among the seven wastes was adapted from Rawabdeh (2005). Each question has a different number of choices, and each choice has a weight ranging from zero to four according to the strength of the relationship. For example, the first question, ‘does n produce m’, has three choices (always = 4, sometimes = 2 and rarely = 0), where n represents any type of waste among the seven lean wastes that affect another type of waste m (see Appendix A). Moreover, a waste assessment questionnaire with 68 items from four categories: man (7 items), machine (12), materials (24) and methods (25) was customised to the apparel industry from Rawabdeh (2005) and used to prioritise the lean wastes. The questions with the note ‘From’ represent existing wastes that may cause another type of waste, while the questions with the note ‘To’ indicate that the existing waste might be caused by other wastes. The number of questions with the notes ‘From’ and ‘To’ are presented in Table 4. The answer to each question has three levels of agreement (yes = 1, medium = 0.5 and no = 0) (see Appendix B). The original waste assessment and waste relationship questionnaires were developed by Rawabdeh (2005) and tested by himself and other researchers (Kosasih et al., 2019; Suhardi et al., 2020).

Table 4. Groups of waste assessment questions by type of waste.

S. No	Type of question	Number of questions (Ni)
1	From overproduction	3
2	From inventory	6
3	From defects	8
4	From motion	11
5	From transportation	4
6	From process	7
7	From waiting	8
8	To defects	4
9	To motion	9
10	To transportation	3
11	To waiting	5
Total number of questions		68

3.4. Data analysis

Suppose that T is an observed turnaround time for a certain operation obtained from a time study. The cycle time (CT) of that particular operation can be obtained as $({\sum }_1^n{t}_i/n)/\mathit{\text{number of resourses}},$ where n is the number of observations taken (n = 30). In addition, lean metrics such as available time = daily working time – relaxation time; $\mathit{\text{Takt time}}=\frac{\mathit{\text{daily available working time}}}{\mathit{\text{daily demand}}};$ lead time = inventory quantities between processes/daily customer requirements; and uptime = percentage of operating time were calculated, and the current state value stream map (CSVSM) was created according to Rother and Shook (2003). Then, the CSVSM was evaluated based on the degree of flow (DF = the ratio of total processing time to lead time) (Erlach, 2013). In addition, from the responses to the waste relationship questionnaire, first, the weights of the answers to the six questions were summed for each relationship with the score indicating the strength of the relationship (1 – 4 = unimportant (U), 5 – 8 = ordinary closeness (O), 9 - 12 = important (I), 13 – 16 = especially important (E) and 17 – 20 = absolutely necessary (A)). Then, for convenience in computation, discrete values were assigned to the letters (X = 0, U = 2, O = 4, I = 6, E = 8 and A = 10) (see Table 6).

Table 5. An algorithm for computing the ranks of the seven wastes (Rawabdeh, 2005).

Steps	Description	Formula
1	Count the number of questions with ‘From’ and ‘To’ notes for each type of waste	Ni (see Table 2)
2	Obtain the weights from the waste relationship matrix, and expand it to the 68 questions following their notes (‘From’ or ‘To’)	See Table 4
3	Divide the values in the table created in step 2 by Ni	$W_{m,k}$
4	Sum the values obtained from step 2 for each waste	$S_m={\sum }_{k=1}^{68}\frac{W_{m,k}}{N_i},$ m stands for each lean waste and k stands for each question
5	Count the frequency of the cells with non-zero weights for each type of waste	Fm
6	Multiply each value obtained in step 3 by the corresponding answers to each question Xk
7	Sum the values obtained in step 6	$s_m={\sum }_{k=1}^{68}{X}_k\frac{W_{m,k}}{N_i}$
8	Count the frequency of the cells in step 6 with non-zero weights for each type of waste	fm
9	Compute the initial indicator Ym for each type of waste	$Y_m=\left(\frac{s_m}{S_m}\right)\left(\frac{f_m}{F_m}\right)$
10	Calculate Pm by multiplying the percentages for the ‘From’ and ‘To’ for each waste type obtained from the waste relationship matrix
11	Calculate the final waste factor Ymfinal	$Y_{\mathit{\text{mfinal}}}=Y_m{P}_m$
12	Rank the seven wastes based on Ymfinal

Table 6. Waste relationship matrix.

To From	O	M	I	T	W	P	D	Score	%
O	A (10)	U (2)	A (10)	I (6)	O (4)	X (0)	I (6)	38	13.7
M	X (0)	A (10)	E (8)	U (4)	A (10)	I (6)	E (8)	46	16.6
I	I (6)	I (6)	A (10)	X (0)	O (4)	X (0)	I (6)	32	11.5
T	E (8)	E (8)	I (6)	A (10)	O (4)	X (0)	E (8)	44	15.9
W	E (8)	X (0)	A (10)	X (0)	A (10)	E (8)	X (0)	36	12.9
P	I (6)	E (8)	E (8)	X (0)	U (4)	A (10)	I (6)	42	15.1
D	U (2)	A (10)	E (8)	I (6)	U (2)	U (2)	A (10)	40	14.4
Score	40	44	60	26	38	26	44	278
%	14.4	15.8	21.6	9.4	13.4	9.4	15.8		100

From the waste assessment, using 68 questions, first, the responses were averaged. Then, by integrating with the results from the waste relationship matrix, the ranks of the seven wastes were computed by the algorithm presented in Table 5 (Rawabdeh, 2005). It presents the necessary steps to be followed while analysing the waste assessment responses to rank the lean wastes. Using the ranks (Table 8), an appropriate VALSAT was selected, and the wastes were analysed (Hines & Rich, 1997). Following the reduction of the main wastes by applying layout improvement and 5S, a time study was conducted after three months to create the future state value stream map (FSVSM) according to the value stream design guidelines provided in Erlach (2013) and Rother and Shook (2003). The lean tools were selected based on literature (Hodge et al., 2011; Kosasih et al., 2019; Manickam & Rathinasamy, 2022).

Table 7. Sample analysis of the responses to the waste assessment questionnaire.

Question (Q.) type	Ni	k	AR	Wo,k	Wm,k	Wi,k	Wt,k	Ww,k	Wp,k	Wd.k
Man (7 Q.s)
To motion	9	1	0.366	0.08	0.41	0.25	0.41	0	0.33	0.41
From motion	11	2	0.167	0	0.15	0.15	0	0.15	0.12	0.12
From defects	9	3	0.567*	0	0.71	0.57	0.71	0.14	0	0.71
From motion	11	4	0.133	0	0.12	0.12	0	0.12	0.09	0.09
Material (24 Q.s)		⋮
To waiting	5	8	0.67	0.53	1.34	0	0.53	1.34	1.07	0.27
From waiting	8	9	0.57	0.57	0	0.57	0	0.71	0	0.14
From transportation	3	10	0.22	0.44	0.55	0.44	0.55	0.22	0	0.44
From inventory	6	11	0.85	1.42	0.85	1.42	0	0	0	1.13
From inventory	6	12	0.92	1.54	0.92	1.54	0	0	0	1.22
From defects	9	13	0.32	0	0.41	0.32	0.41	0.08	0	0.41
From inventory	6	14	0.9	1.5	0.9	1.5	0	0	0	1.19
From waiting	8	15	0.63	0.63	0	0.63	0	0.79	0	0.16
To defects	2	16	0.8	1.6	1.6	1.6	1.6	0.4	1.2	2
From defects	9	17	0.56	0	0.7	0.56	0.7	0.14	0	0.7
Machine (12 Q.s)		⋮
From process	7	32	0.3	0.17	0.34	0.43	0	0.34	0.43	0.26
To waiting	5	33	0.67	0.54	1.34	0	0.54	1.34	1.07	0.27
From process	7	34	0.2	0.11	0.23	0.29	0	0.23	0.29	0.17
From transportation	3	35	0.23	0.47	0.59	0.47	0.59	0.23	0	0.47
Method (25 Q.s)		⋮
To transportation	3	44	0.62	1.24	0	0	2.06	0	0	2.06
From motion	11	45	0.2	0	0.18	0.18	0	0.18	0.14	0.14
From waiting	8	46	0.7	0.7	0	0.7	0	0.875	0	0.175
		⋮
From motion	11	65	0.133	0	0.12	0.12	0	0.12	0	0.09
Form overproduction	5	66	0.52	1.7	0.35	1.39	1.04	0.69	0	1.39
From process	7	67	0.23	0.13	0.26	0.33	0	0.26	0.33	0.19
From defects	8	68	0.56	0	0.7	0.56	0.7	0.14	0	0.7
Score (sm)				37.5	36.2	38.6	31.5	23.2	17.9	43.7
Frequency (fm)				50	57	60	37	50	33	68

Note: W_o,k: weight of overproduction waste according to k^th question (k = 1, 2, 3, … 68); AR: the average of the 30 responses for each question, k = 1, 2, 3, …, 68, *highlighted values show the questions (wastes) with higher scores (near to 1 which is a complete agreement of its existence).

Table 8. Summary of lean waste assessment and the rank of the seven lean wastes.

	O	M	I	T	W	P	D
Sm	60.9	74	75.8	58.9	49.6	38	82
Fm	50	57	60	37	50	33	68
sm	37.5	36.2	38.6	31.5	23.2	17.9	43.7
fm	50	57	60	37	50	33	68
Ym	0.615	0.489	0.509	0.53	0.467	0.47	0.53
Pm	197.28	262.28	248.4	149.4	172.86	141.94	227.52
Ym final	121.32	128.25	126.43	79.18	80.69	66.7	120.58
Final result (%)	16.78	17.7	17.5	10.95	11.2	9.2	16.67
Rank	3	1	2	6	5	7	4

The analysis of the waste relationships and waste assessment responses were held in Microsoft Excel and the value stream maps were created using ConceptDraw software version 9. The FSVSM intends to create a lean value stream by linking the final customer back to the raw material smoothly which results in the shortest lead time and lowest cost by maintaining acceptable quality. To design the FSVSM, the following steps were followed (Rother & Shook, 2003):

1. Calculate the takt time and aim to produce up to it

2. Develop a continuous flow wherever possible

3. Use supermarket pull system wherever continuous flow doesn’t work

4. Select the pacemaker process

5. Level the production mix

6. Level production volume

7. Identify the required improvements

8. Iterate the process continually where necessary

Finally, comparisons of the current and future state value stream maps were performed based on cycle times (Reda & Dvivedi, 2021; Yadrifil et al., 2020), standard minute values and work-in-process levels (S. Kumar & Thavaraj, 2015) and lead time: the most common performance measures (Romero & Arce, 2017). Takt time, the rate at which a company must produce a product to satisfy its customer demand (B. Singh & Sharma, 2009), is commonly used as a target (Reda & Dvivedi, 2021; Rohani & Zahraee, 2015). While there are studies that further improved the takt time (Deshkar et al., 2018; Zahraee et al., 2020, 2021), improving the takt time was not within the scope of the current study but was considered as a target. The standard minute values were computed as follows (Suhardi et al., 2020):

The normal time (NT), the time a work is completed by a worker in reasonable conditions = Observed processing time*Rating factor (depends on operator performance).

The standard minute value (SMV) = NT (1+A), where A = allowance which is added to be realistic and achievable. In this study, the operator rating factor and allowance are taken to be 0.8 and 0.3, respectively.

4. Case study

4.1. Company profile

The case company is a private limited garment manufacturing company founded in Saris industrial zone, Addis Ababa, Ethiopia operating since 2011 with around 343 employees. The company is known for producing dress shirts, neckties, boxer pants and knitwear for the local and export market based on a make-to-order production strategy. The attainable yearly production capacity of dress shirts (men and women), boxer pants and knitwear are 453,600 pieces, 9600 pieces and 9600 pieces, respectively (Ethiopian Textile Industry Development Institute, 2021). The study was conducted on a production floor of a specific men’s shirt model. The production floor of a shirt manufacturing process was selected because studies show that sewing room efficiency in the Ethiopian garment industry is higher for basic products such as T-shirts than for complicated garments such as shirts (Abtew et al., 2020), and the operational cost of a shirt is high (S. Prasad & Panghal, 2022). It consists of 45 operations including cutting (5 operations), collar preparation (9), cuff preparation (4), sleeve preparation (3), front preparation (7), back preparation (3), assembly (10) and finishing (4). The company works six days per week with one shift of 8 h (excluding 1-h lunchtime) per day. Deducting a 15-min relaxation time, it has a working time of 7 h and 45 min (29,700 s) per day. The production floor consists of one cutting section, a preparation section, three assembly lines and a finishing section.

4.2. Current state value stream map

To create the CSVSM and analyse the existing situation of the case company, a weekly order of 3000 pieces (428 pieces per day) was considered. The customer takt time (TT) was calculated to be 65.19 s per piece (29,700 s/428 pieces). The number of operators in each section, cycle time, available time (AT), uptime (UT), inventory between processes and lead times are depicted in Figure 2. The degree of flow is about 0.58% which shows that more than 99% of the lead time is non-value-adding time. The cycle times of the aggregated four main sections with several sub-processes are quite above the customer takt time. The cutting and preparation sections hold high inventories resulting in waiting time waste in subsequent sections because of capacity imbalance. Long distances between the machines used to prepare components such as fusing, collar and cuff turning and sharpening increase non-value-adding times arising from the poor layout. The ironing machines were located only in the finishing section leading to the movement of bundles for ironing which interrupts processes and creates motion, transport and waiting wastes. High work-in-process inventories create a cluttered workplace and longer lead times. These results indicated a potential for improvement. Besides the metrics in the CSVSM, understanding the relationship between the seven lean wastes and the ranks of lean wastes from the waste assessment was used to select an improvement scheme.

Figure 2. Current state value stream map of a shirt manufacturing process.

4.3. Lean waste identification and prioritisation

Table 6 presents a waste relationship matrix, in which motion (16.6%), transport (15.9%) and over-processing (15.1%) are found to be the main wastes influencing all other wastes. Improper layout caused an increased motion of workers and transport of raw and work-in-process materials. The increased material handling creates defects causing over-processing. The result is in agreement with Suhardi et al. (2020) and Rimawan et al. (2018), while it differs from Henny and Budiman (2018) and Amrina and Andryan (2019) who found defects as the most influential wastes. Table 6 also shows that inventory (21.6%), defects (15.8%) and motion (15.8%) are the most influenced wastes by other types of lean wastes which concurs with Henny and Budiman (2018) and Rimawan et al. (2018).

A further assessment of waste was conducted to determine the main lean wastes. The average responses (AR) of the four categories of the questions showed that materials (AR = 0.571), machines (AR = 0.481), methods (AR = 0.476) and man (AR = 0.319) are the sources of waste in decreasing order. Poor inventory management, work-in-process materials staying long in process areas, materials moving more often than necessary and cluttering in receiving areas of sections, and delayed material requests to stores are wastes arising from materials. Poor and limited material handling equipment and unpredictable machine workloads are wastes from machines. Lack of clear procedure for inspection, purchasing plans not considering items available in-store, inflexibility of production times, inefficiency in stores and delays not communicated are the main wastes from methods (for example, see the highlighted values on the AR column of Table 7).

Tables 7 and 8 present a sample and summary of the analysis of the responses to the waste assessment questions, respectively. The results in both tables are computed based on the algorithm presented in Table 5. As can be seen from the ranks of the lean wastes in Table 8, motion is found to be the major type of waste followed by inventory, overproduction and defects. Observation of the researchers indicated that operators mostly moving from machine to machine to perform some operations result in increased material handling, opportunities for loss of components, creation of defects, scraps and reworks. The labour intensiveness of the apparel industry leads to the enlarged effect of motion waste and factory disorganisation, especially in emerging countries such as Ethiopia where labour skill is very low, and there is an ergonomic problem in workstation designs (Mulugeta, 2021). Increased inventory was found to be the second major type of waste. Accumulation of inventory leads to longer lead times affecting the performance of the apparel industry where meeting an agreed delivery time is vital in satisfying customers and maintaining success in the market. The progressive bundle production system also contributes to increased inventories and bundle mix-ups. Overproduction, the third major waste, happens when the companies make garments to stock in between orders. Moreover, the overproduction of cut pieces creates a burden on preparation (a section right after cutting) and subsequent sections. Defects arising from operators who are not well-trained and stay for a short period in the company due to high turnover (Taplin et al., 2003) is the fourth waste, while transportation was not noticed by the employees. However, the diagnosis of the existing layout showed considerable transportation waste due to improper layout.

4.4. Improvements and the future state value stream map

The process activity mapping (PAM) tool was used to analyse the wastes in the case company. It was selected due to its highest score among the seven VALSATs (Hines & Rich, 1997). PAM grouped the production activities into three: VA = 1548.04 s (0.57%), NNVA = 20,373.93 s (7.56%) and NVA = 247,502.02 s (91.86%). The result shows that NVA activities are the major part of the production lead time. Figure 2 illustrates that the assembly line had a better performance compared to the preparation and cutting sections. These two sections suffer from poor layouts and cluttered materials. Therefore, layout improvements and 5S (sort, set in order, shine, standardise and sustain) were implemented to reduce the main wastes: motion, inventory and overproduction.

4.4.1. Layout improvement

First of all, the arrangement of the cutting, preparation, assembly lines and finishing sections with respect to common machines such as fusing, sharpening, ironing and overlock sewing machines was not appropriate. Hence, a product-based suitable layout for the selected long-sleeve shirt was implemented by changing the position of the cutting and preparation sections and bringing them closer to the common machines without affecting continuous flow. This improvement decreased the distances between machines such as cutting to fusing, fusing to finishing, preparation section to collar and cuff turning machine by 75, 38 and 46%, respectively. As a result, the motion of operators, transport of materials and loss and mix-up of components were reduced. With the new arrangement of workstations, only the sewing machines can be rearranged when there is a style/product type change while the overall arrangement of workstations remains unchanged.

4.4.2. 5S implementation

After the arrangement of workstations has been improved by layout improvement, 5S was applied in raw material and finished goods stores and cutting and finishing sections so that only the necessary items were sorted, set in order and shined within the workstations. The implementation was held by a team of five members and the co-author of this paper. A Kaizen expert trained the operators of the different sections separately. Following the conducive environment created after the training, the team implemented 5S step-by-step following the plan, do, check and act cycle starting from the raw materials store and scaling it up to the other sections. To standardise and maintain sustainability, labelling of materials was applied so that one can easily identify changes and take corrective actions.

4.4.3. Pull system and FIFO

The high work-in-process inventories in the cutting, preparation and finished goods warehouse were reduced by implementing the concept of producing only when the subsequent section or the end user places an order. Hence, cutting could no longer accumulate cut pieces because of its excess capacity, instead, the number of operators was decreased, and cutting was performed when needed. The FIFO rule was applied by moving bundles based on their sequence through bundle tickets written on fabric swatches. Moreover, decreasing the bundle size and the number of bundles in a basket reduced the piles of work-in-process inventories between sections and overproduction and enabled the operators to detect defective components earlier which reduced delayed reworks and scraping of garments.

4.4.4. Future state value stream map

The implementation of the lean tools reduced the main wastes: motion, inventory and overproduction. To demonstrate the improvements, the FSVSM was created by conducting a time study after the implementation of the mentioned lean tools and monitoring the system for three months. Figure 3 depicts the FSVSM with the implemented lean tools and lean metrics including: the number of operators, work-in-process inventories, cycle time, processing time and lead time.

Figure 3. Future state value stream map of a shirt manufacturing process.

5. Results, discussion and implications

The VSM along with WAM and the two lean tools enabled the researchers to look at lean wastes from different perspectives including the overall existing wastes (average response = 0.5 indicating that the respondents agreed on the presence of waste on a medium level), the main sources of waste (materials and machines) and the relationship (Table 6) and ranks of the seven wastes (Table 8). Motion, inventory, overproduction and defects were found to be the major wastes in their decreasing order. Rawabdeh (2005), Rimawan et al. (2018) and Suhardi et al. (2020) also found motion as the major lean waste in steel manufacturing, assembly of vehicles and the apparel industry, respectively. However, Behnam et al. (2018) found defects as the priority waste and motion as the last priority in clothing manufacturing because no layout problem was observed in their study.

This understanding created good opportunities for improvement. Figures 4–6 present comparisons of the current and future states in which promising improvements have been achieved. From Figure 4, it can be seen that considerable WIP has been reduced in each production section with the highest reduction of 86.38% being achieved in the finishing section. The overall average WIP reduction accounts for 74.30%. Figure 5 depicts the improvements in value-added time and production lead time. As a result of the lean implementation, the value-added time and production time have been reduced by 30.95 and 71.10%, respectively. As can be seen in Figure 6, the SMV of the shirt has decreased by 27.02%. These improvements are attributed to the layout improvement which resulted in a reduced motion of operators and transport of materials, and the pull system and FIFO which decreased inventories and overproduction.

Figure 4. WIP inventory of shirt manufacturing processes.

Figure 5. Value-added time and production lead time of shirt manufacturing.

Figure 6. SMV of shirt manufacturing.

5.1. Discussion

By integrating waste assessment with the implementation of a bundle of lean tools along with VSM, the performance of the shirt production process has been improved. Compared to the current state, the production lead time decreased by 71.10% in the future state. According to a review of the application of VSM in manufacturing industries (Romero & Arce, 2017), the average improvement of lead time reported in 54 papers is 52.26%. Our result is also better than the lead time improvements recently reported by Reda and Dvivedi (2021) (69.70%) and Zahraee et al. (2021) (36.36%). Our result indicates an average WIP reduction of 74.30%. This is better than the average WIP reduction reported by Romero and Arce (2017) (64.75%) and S. Kumar and Thavaraj (2015) (72.97%). Both improvements are substantial because WIP inventory and production lead time are interrelated and the case company was underperforming before the conduction of this study.

Moreover, the value-added time has been reduced by 30.95% which is better than that of J. Singh et al. (2020) (14.71%) but lower than the improvement reported by Deshkar et al. (2018) (74.85%). The SMV of the shirt decreased by 27.02% which is comparatively lower than the SMV improvement reported by S. Kumar and Thavaraj (2015) (36.25%). However, these results can further be improved through time (Wickramasinghe & Wickramasinghe, 2017) to make the production process leaner. The leaner the production process, the lesser the production cost and the better the product quality becomes (Carvalho et al., 2019).

5.2. Implications

Here, theoretically, it is worth mentioning that identification and prioritisation of the lean wastes in the context of the company is necessary before picking any lean tool for implementation. The results of our study demonstrated that the combined application of WAM and VSM along with layout improvement and 5S reduced the main wastes: motion, inventory and overproduction. The application of VSM in the apparel industry can be expanded to different products and styles (Silva, 2012) which have been produced by a company repeatedly as VSM has limitations in mapping multiple products (Sultana & Islam, 2013). The method provided here can also be extrapolated to other low-level technology industries following a similar procedure.

Practically, as production managers in the apparel industry in Ethiopia are challenged to meet delivery times, with proper factory settings such as layout and 5S implementation, garment factories can reduce waste and minimise delivery delays (Obeidat et al., 2012). Eliminating or reducing lean wastes have an eventual effect on the sustainability of the apparel industry (Phuong & Guidat, 2018). The findings of this study have implications for the managers of apparel manufacturing companies in Ethiopia to consider collective lean tools along with VSM to visualise and eliminate NVA activities. In emerging economies, where resource is limited, the opportunity cost of NVA activities is high. Lean tools, with minimum investment and continual small improvements, are suitable for the apparel industry and can save resources to be used in other value-adding practices and boost profit. Hence, the fragmented implementation of Kaizen in several industries (Georgise & Mindaye, 2020) and some lean tools in the apparel industry (Mulugeta, 2021) in Ethiopia need to be managed in an institutional way (Suárez‐Barraza & Ramis‐Pujol, 2012) to reduce waste, improve profitability and support the transformation of the country’s economy.

6. Conclusions

This case-study-based research was conducted with the aim of implementing a VSM technique along with the WAM and two other lean tools in a shirt manufacturing company in Ethiopia and demonstrating its applicability. The data were collected through time studies and two structured questionnaires. The findings showed that motion, inventory and overproduction are the first three main wastes. Process activity mapping showed that about 91.86% of the production lead time was NVA activities. Implementation of layout improvement and 5S resulted in transforming the old state into the new state with significant reductions in the number of operators, value-added time, WIP inventory, production lead time and SMV of a shirt. The findings suggest that managers of garment manufacturing companies in Ethiopia should look at their production floors, initiate lean production systems, create awareness, allocate resources and train their employees to eliminate wastes. Observation of the researchers indicate that the employees are willing to learn new ways of doing their jobs.

6.1. Limitations and future study

Despite the efforts made before the lean implementation through the survey instruments to identify and prioritise the lean wastes and the significant improvements achieved, the three-month period of the actual implementation is short. Moreover, the sample size for the survey was small. This was because the questions are many, and few employees have an overall understanding to respond to the entire questionnaire. Focus group discussion was not possible because of a failure to find an agreeable common time. This study was limited to a shirt manufacturing company. Otherwise, if the procedures are properly followed, VSM, WAM and the implemented lean tools can be applied to any setting. However, the main lean wastes and the appropriate lean tools to be used may differ from industry to industry.

Future case studies should consider a longer period of implementation time and focus group discussions to get more significant improvements or the integration of simulation with VSM and the improvement of critical lean metrics such as takt time. Moreover, sustainable VSM can be utilised to improve the triple bottom lines: economic, environmental and social sustainability of industries.

Acknowledgements

The authors are thankful to the Ethiopian Institute of Textile and Fashion Technology, Bahir Dar University for covering the expenses of the study through the fund.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Funding

This work was conducted under the centre of excellence for the higher education and TVET program Ethiopia-phase 3, PE479 higher education, KFW project number 51235 and BMZ number 201166305.

Notes on contributors

Berihun Bizuneh

Berihun Bizuneh is an Associate Professor at the Ethiopian Institute of Textile and Fashion Technology, Bahir Dar University, Bahir Dar, Ethiopia. He holds a PhD in Industrial Management from the National Taiwan University of Science and Technology (Taiwan, 2019), an MBA in Marketing from Mekelle University (2011) and a BSc in Textile Engineering from Bahir Dar University (2007). He has broad experience in research on industrial management with a focus on production management, quality control, and data mining applications. Berihun possesses teaching experience in textile economics, quality control, product development, project management, operations management, and operations research.

Rahmet Omer

Rahmet Omer is a Lecturer at the Department of Garment Engineering, Wolkite University, Wolkite, Ethiopia. She holds an MSc in Fashion Technology (2022) and a BSc in Garment Engineering from Bahir Dar University, Bahir Dar, Ethiopia. She has teaching experience in lean manufacturing systems, cutting technology, pattern making, garment construction, high-performance products, production planning and control, apparel trim and accessories, and statistical applications in the garment industry.

Appendix A.

Waste relationship questionnaire adapted from Rawabdeh (2005)

Table

No	Question	Choices	Weight
1	Does n produce m?	Always Sometimes Rarely	4 2 0
2	What is the type of relationship between n and m?	As n increases, m increases As n increases, m reaches a constant level Random (depends on conditions)	2 1 0
3	The effect of m due to n:	Appears directly and clearly Needs time to appear Not often appears	4 2 0
4	Eliminating the effect of n on m is achieved by:	Engineering methods Simple and direct methods Instructional solution	2 1 0
5	The effect of m due to n, mainly influences:	Quality of products Productivity of resources Lead time Quality and productivity Productivity and lead time Quality and lead time Quality, productivity, and lead time	1 1 1 2 2 2 4
6	To which degree does the effect of n on m increase manufacturing lead time?	High degree Medium degree Low degree	4 2 0

Note: n stands for any type of waste which affects the other type of waste m.

Appendix B. Waste assessment questionnaire customised from Rawabdeh (2005) (Y = yes = 1, M = medium = 0.5, and N = no = 0)

SN	Category	Type	List of questions	Y	M	N
1	Man	To motion	Management transfer operators to various jobs and machines so that operations can be performed by all individuals
2		From motion	Supervisors frequently check operators but they don’t empower them
3		From defects	Hourly workers are not adequately supervised during overtime
4		From motion	Positive steps are not taken to raise hourly workers’ morale and work interest
5		From motion	Training for new employees is usually provided on the job
6		From defects	Workers don’t instil a sense of responsibility towards their jobs
7		From process	Workers’ safety protection tools are not used in the work area
8	Material	To waiting	Supplier’s lead time is not usually made available to production schedulers
9		From waiting	Schedules are not usually checked for material availability before being released to production
10		From transportation	Components are not received in unitised loads/appropriate bundle sizes
11		From inventory	Production planning doesn’t give sufficient advanced notice of required items to warehouse personnel
12		From inventory	Warehouse personnel is not notified in advance about planned inventory changes
13		From defects	There is an excessive accumulation of materials awaiting to be repaired, reworked, or returned to suppliers
14		From inventory	Materials appear to be standing around piled up unnecessarily on the receiving platforms of different sections
15		From waiting	Production workers usually stand around waiting for materials to arrive
16		To defects	Materials are moved more often than necessary
17		From defects	Small components are frequently damaged, mixed, or lost in transportation activities
18		From transportation	Items in temporary storage areas are mixed with the materials being used or the finished products which need to be transferred to the next process
19		To motion	Loading and unloading of raw materials or components are not handled in an appropriate manner
20		From waiting	Adequate containers are not used before the packaging process to simplify the counting process and make the transfer of goods easier
21		From motion	Similar goods or raw materials are not stored in one area to facilitate and reduce the time required for searching
22		From transportation	There are repeated transfers of goods or raw materials with small containers
23		From defects	Incoming materials are not properly checked to determine the suitability of their standards for the quality and quantity of goods
24		From motion	Material or goods are not labelled for the convenience of identification
25		From inventory	Workers keep goods that are in process (WIP) in the production process areas
26		From inventory	Standard orders/ materials are stored in warehouse supplies, although not needed immediately
27		To waiting	There is free time for items that are still inside process (WIP) before further processing
28		From defects	There is a rework process for size/weight/shape/product color mismatch
29		From waiting	Raw materials don’t usually arrive on time when needed
30		From overproduction	There is an accumulation of raw materials in the warehouse when there are no customers scheduled
31		To motion	Raw materials and equipment are not properly stored
32	Machine	From process	Tests on the efficiency of the machines, compared to the manufacturer’s specifications, are not periodically conducted
33		To waiting	The workload for each machine is not predictable in sufficient detail
34		From process	Once a machine is installed, there is no sufficient follow-up to see if it performs according to specifications
35		From transportation	The capacity of the material handling equipment /transportation can is not enough to carry the heaviest items/bundles
36		To motion	When using moving tools for goods or raw materials, the amount of material being carried is not full
37		From overproduction	There is a management policy to produce more than necessary to maximise capacity and machine usage
38		From waiting	The machines stop frequently due to mechanical disturbances
39		From waiting	The necessary tools are not always available and sufficient for the production process of each product
40		To defects	The material handling equipment is at risk of damaging materials or products
41		From waiting	Long setup times may cause delays to the operating flow
42		To motion	There are tools that are damaged or unused in the work area
43		From process	The considerations for reducing machine setup time by customising scheduling and design are rarely practiced
44	Method	To transportation	The storage area is insufficient to avoid the blockage of warehouse aisles
45		From motion	There is no labelling for materials to make it easier to store, refer and retrieve raw materials or products
46		From waiting	The storage space is not used effectively for storage with the aid of shelves and trolleys
47		To motion	There is no division of warehouse area, activity areas for orders most frequent, and reserve areas for other orders
48		To waiting	The production time is not adjusted to the number of needs and customer orders
49		To defects	The production schedule is not well-communicated to all sections, and its contents are not widely understood
50		From motion	The standard operating procedure to use machines in the production process is rarely adhered
51		From defect	There is no quality control system for each department/section to guarantee quality
52		From motion	There is no preset standard time for each operation
53		To waiting	When there is a delay, it is not well-communicated to all sections
54		From process	There is no schedule arrangement to avoid repeated machine settings/arrangements to reproduce similar products
55		From process	The possibility to combine steps so that processing becomes simpler is rarely practiced
56		To defects	There is no procedure for process inspection or inspection against returned products
57		From inventory	The inventory records are not used to make purchases of raw materials and production schedules
58		To transportation	The aisles are not well cleaned and tidied up
59		To motion	The storage areas are not marked and labelled in certain sections
60		To transportation	The alley area is not sufficient for transportation movement with freedom
61		To motion	There is improper storage of materials stored in the warehouse area
62		To motion	There is no regular schedule for cleaning the production area
63		From motion	The production doesn’t flow in one direction
64		From motion	There is no team that handles designs, sample development, and standardisation of processing time
65		From motion	Work standards don’t have clear and specific goals
66		From overproduction	Work imbalances cannot be forecasted in time to adjust them
67		From process	The working procedures are not capable of screening or eliminating redundant or unnecessary works
68		From defects	The result of quality control, product testing, and evaluation are not passed on to engineering for improvement