Lean-based clean earthworks operation

Journal of Cleaner Production 142 (2017) 2195e2208

Contents lists available at ScienceDirect

Journal of Cleaner Production journal homepage: www.elsevier.com/locate/jclepro

Lean-based clean earthworks operation lez b, Tak Wing Yiu b Sheila Belayutham a, *, Vicente A. Gonza a b

Faculty of Civil Engineering, Universiti Teknologi MARA, 40450, Shah Alam, Selangor, Malaysia Department of Civil and Environmental Engineering, University of Auckland, 20 Symonds Street, 1010, Auckland, New Zealand

a r t i c l e i n f o

a b s t r a c t

Article history: Received 7 July 2016 Received in revised form 10 November 2016 Accepted 10 November 2016 Available online 12 November 2016

Earthworks operation occupies only a short period of the total project duration but comes with a high cost, mainly due to the use of heavy machineries and skilled operators. Regardless of the short duration, negative effect of the operation on the environment is detrimental, especially from the perspective of site sediment pollution. However, the current body of knowledge lacks improvement strategies that could enable simultaneous enhancement of the production and environmental factors during the operation period. Due to that, this study aims to seamlessly improve the production (time and cost) and environmental (site sediment pollution) variables by applying the concept of lean production towards achieving a cleaner earthworks operation. As a result, a lean based methodology has been proposed by using case study approach that combines different data collection methods (interview, observation, site document). Findings of the study suggest that lean enables clean. Positive improvements have been observed in terms of time and cost reduction by 42.7% and 24.9% respectively. The environmental factor, Rainfall Erosivity is reduced by 41.8%, consequently reducing the risk of soil erosion and sediment production. Ultimately, the proposed methodology could seamlessly improve both dimensions of production and environment at its source, which satisfies the aim for cleaner earthworks operation. © 2016 Elsevier Ltd. All rights reserved.

Keywords: Lean production Cleaner production Lean-clean Earthworks operation Sediment pollution Exploratory research

1. Introduction Earthwork operation takes place during the early stages of construction where it involves land clearing and grading for a short period of time. Regardless of the short duration of occupation, the environmental threat is detrimental, especially from the aspect of water pollution with sediment as the pollutant (Ooshaksaraie et al., 2009; Taylor and Field, 2007). Therefore, earthwork is a critical work stage that requires proper management because an uncontrolled cleared site could result in sediment pollution (Brown and Caraco, 1997). Sediment pollution could create chains of other problems such as damage to the aquatic ecosystem, health risk to the people and unnecessary cost and resources for remedial works (Harbor, 1999). Conventionally, the risk of sediment pollution during construction has been kept under control using end-of-pipe approaches by allocating erosion and sediment control facilities, e.g. mulches and sediment pond (NZTA, 2010). Those control facilities do come with additional cost and time to be installed (Shaver, 2000). Besides the downside on the production variables of

* Corresponding author. E-mail address: [email protected] (S. Belayutham). http://dx.doi.org/10.1016/j.jclepro.2016.11.060 0959-6526/© 2016 Elsevier Ltd. All rights reserved.

time and cost, treating the already produced pollutant is supposed to be the last approach in terms of environmental management (Hamner, 1996). Control at source concept such as Cleaner Production (CP) proposes an alternative approach to manage environmental issues by preventing or limiting the occurrence of the pollutant (UNEP, 1996). CP functions to increase productivity through the efficient use of resources while promoting better environmental performance through source reduction and emission (Kjaerheim, 2005; Cabello Eras et al., 2013). Even though CP has provided the basis for improving both dimensions of production and environment, the environmentally driven concept still lacks prescriptive techniques to address the operational inefficiencies in production. It is common for CP to rely on technological changes (Neto et al., 2013) but fundamentally, CP is not all about technology and should start from the basis of improving the current processes through elimination of its operational inefficiencies. On the other hand, LP has been established primarily to improve the performance of operations from the production aspect and has been long-practiced to improve the production processes in various sectors (Womack et al., 1990). Lean has also been proven to benefit in terms of the environment whether as a coincidence or planned benefit. In regards to that, lean is believed to enhance clean (Cobra et al., 2015; Queiroz et al., 2015).

2196

S. Belayutham et al. / Journal of Cleaner Production 142 (2017) 2195e2208

1.1. Research aim and objective This study aims to adopt the principles of Lean Production into the concept of Cleaner Production in order to improve the production (time and cost) and environmental (site sediment pollution) variables of earthworks operation. In this study, the term Cleaner Production will be used interchangeably with ‘clean’ while the term Lean Production will be used interchangeably with ‘lean’. The objective of this study is to develop a practical lean-based methodology to enhance the functions of clean in an earthworks operation with the environmental focus on site induced sediment pollution. For this study, the developed methodology will be known as lean-clean. 2. The production and environmental issues of earthworks operation The production factor of time has been hailed as a crucial variable in earthworks because this operation sets the rhythm for subsequent activities (Fu, 2013). The operation also acquires a relatively high cost in comparison to the time spent on the work, due to the heavy dependence on machineries and skilled operators (Kang et al., 2009). Hence, it is common for productivity to be the subject of interest among the industry players and researches who seek to improve the earthworks operation (Martinez, 1998; Dawood et al., 2010). Previously, the productivity studies have focused on the traditional aspects of construction which are time, cost and quality without relating it to the environment. However, the current scenario differs due to progressions that have been made towards integrating environment as part of the studies conducted to improve the performance of earthwork operations. For example, Golzarpoor et al. (2013) have provided a synergistic approach that combines the production and environmental factors in determining the cost, fuel and energy usage and lez and carbon emission from the earthwork operations. Gonza Echaveguren (2012) and Capony et al. (2012) have also conducted similar research using discrete event simulation and GPS technology respectively. However, most of the studies have concentrated on the common issues of air and carbon emission with least regards for site sediment pollution. Typically, the environmental issue of site sediment pollution has long been treated in isolation from the production aspect (Lewis and Hajji, 2012). According to Belayutham lez (2015), the independent treatment of the aforeand Gonza mentioned variables of production and sediment pollution will cause contractors to exert imbalance efforts in addressing the different variables. In this case, the environmental issue is commonly being side-lined. Hence, this study intents to identify a common ground between the production and environmental (site sediment pollution) variables of earthworks operation that allows both aspects to be managed seamlessly through the adoption of Lean Production. 2.1. Lean production and earthworks operation Lean Production (LP) is a manufacturing-based production management philosophy that has been applied in construction with the term lean construction (Koskela, 2000). The application of lean in earthworks for construction projects could be categorized as pure lean or technologically infused lean-approaches (Belayutham lez, 2015). For pure lean approaches, Fidler and Betts and Gonza (2008) and Kaiser and Zikas (2009) have used lean tools and principles to stabilize and improve the efficiencies of the earthwork's movement, increase equipment utilization, cost reduction and optimize labor resources. For improvements done with the help of technology, Dawood et al. (2010) have produced an

interactive visual lean system for earthwork operations planning to achieve transparency, reduce complexity, waste and positive project time. Similarly, Kemppainen et al. (2004) have used two optimization algorithms to assist in finding the most cost-efficient schedule and mass haul alternatives that ultimately increased the functions of the Last Planner System in Finland's construction industry. Meanwhile, Kirchbach et al. (2014) have presented ‘digital kanban’, a system supported by machine sensory and information technology that embraces lean principles for optimized earthwork productivity. Most works have been done to apply lean to improve the earthworks' production factor, with little effort found to enhance the environmental variable, specifically sediment pollution. Nonetheless, LP has been combined with green in recent years as the focus on sustainability soars within the construction industry. The relevance has been established as the environmental impact of production processes could be traced back to its inefficiencies (Cabello Eras et al., 2013). The call to integrate lean with the environment has been intense with researches from different industries advocating for the move (Bergmiler and McWright, 2009; Martínez et al., 2009; Lapinski et al., 2006). However, the green-lean integration might not necessarily address the issue of ‘source reduction’ or prevention per se as the term green represents a more general perspective of managing the environmental effect (Baines et al., 2012). As an example in construction, green construction could involve strategies such as using energy-efficient equipment and recycling of waste that has already been produced (Govenor, 2008), rather than to prevent the occurrence of waste at the first instance. In addition to that, it is rather difficult to distinguish a specific definition for green as the term is broad and could relay different meaning to different person (Zaini and Endut, 2014). Therefore, this study intents to utilize a specific ‘control at source’ environmental concept, which is Cleaner Production (CP), that has a clear definition of scope for application so that an orderly method of implementation could be proposed. 2.2. Cleaner Production and earthworks operation Cleaner Production (CP) is “the continuous application of an integrated preventive environmental strategy applied to processes, products and services to increase the overall efficiency to reduce risks to humans and the environment” (UNEP, 1996). CP functions to increase production's productivity through efficient use of the resources and to promote better environmental performance through source reduction (Kjaerheim, 2005; Cabello Eras et al., 2013). The emphasis is to prevent the production of pollution rather than to depend on the end-of-pipe systems where pollutants are being treated after it has been produced (Huisingh and Bass, 1991). A bibliographic study by Giacchetti and Aguiar (2015) on the term CP using the Scopus database found that journal with significant prevalence on this subject is the Journal of Cleaner Production. Therefore, literature search for CP in regards to earthworks operation has been conducted in the CP focused journal. The search in the aforementioned journal using the term ‘earthworks’ has resulted in 43 returns. Within the 43 results, only 2 articles (Belayutham et al., 2016a, 2016b) have discussed particularly on the environmental aspect of site induced water pollution (sediment pollution). From the articles, only 1 article has integrated the use of lean towards cleaner earthworks operation. However, the article has only focused on the off-site related administrative inefficiencies (Belayutham et al., 2016a). Besides that, most of the other earthworks related studies have focused on other environmental variables such as carbon emission (Trani et al., 2016), greenhouse gas emission (Barandica et al., 2013), energy consumption (Cabello Eras

S. Belayutham et al. / Journal of Cleaner Production 142 (2017) 2195e2208

et al., 2013) and construction material waste (Magnusson et al., 2015). Therefore, this study intents to fill the current gap of knowledge by providing a lean based methodology towards achieving a cleaner earthworks operation that focuses on the environmental aspect of site sediment pollution. 2.3. Collation between lean production and Cleaner Production CP is an environmental management concept that emphasizes in making changes to the operational practices in order to have a cleaner operation that minimizes environmental damage, simultaneously increasing the efficiency of the production processes. Nonetheless, the methodology to address the ‘how’ to make the changes to the current operation practices is quite scarce and vague. Generally, Geiser (2001) has mentioned that CP has progressed much from its initial establishment but much more is required to be done, especially in regards to the deficient use of common metrics to measure performance. Queiroz et al. (2015) have found a strong synergy between LP and CP whereby many of the LP tools can positively enhance the environmental aspect in a system. According to Cobra et al. (2015), even though CP and LP present much similarity, it is still a challenge to design an effective integrated approach that could satisfy both the practical and empirical demand. Additionally, the integration between LP and CP is still at its infancy as both concepts have been conceived for different reasons (Degani and Cardoso, 2002). This is evident as the search using the keyword ‘clean lean’ in the Journal of Cleaner Production has resulted in only 1 article by Belayutham et al. (2016b). A search in the Google Scholar has provided an article by Wu and Low (2012), who have demonstrated the benefits of the synergy between LP and CP. Lean principles that have been applied in the study have promoted the search for ineffective and inefficient activities that does not entail high investment cost. By analyzing the flow and energy consumption in an organization, avenues to identify emissions using source reduction strategies in the production process could be done. Due to the strength of lean in production improvement, it could be used to increase the functionalities of clean that could benefit the production and environmental dimensions with the emphasis on ‘source reduction’. Therefore, the lean-clean concept should go hand-in-hand to improve both the production and environmental dimensions (Degani and Cardoso, 2002). 3. Research method 3.1. Exploratory research Exploratory study is conducted when researchers found minimal or no scientific knowledge on the subject under study, which is worth exploring due to the possible benefits from the discovery (Stebbins, 2001). In that respect, this study is an exploratory research due to the limited studies found to simultaneously improve the operational and the environmental aspect of earthworks, particularly site induced water pollution. In terms of sediment pollution during earthworks, most of the green and environmental sustainable related works focused on mitigation systems which are designed at the planning stage of a project. Minimal work is found to improve site induced sediment pollution in tandem with the operational improvements. Exploratory study is different compared to a confirmatory study because the former is more concerned in using descriptive statistics rather than inductive statistics (Jaeger and Halliday, 1998; Stebbins, 2001). Descriptive statistics focuses on gathering and displaying data into information such as indexes, percentages, and frequency distributions, rather than to generalize the knowledge (Cleef, 2014).

2197

The data collected could be both quantitative and qualitative data. Exploration and inductive reasoning are important in science because deductive reasoning on itself could not reveal new ideas and observations (Stebbins, 2001). Hence, this exploratory based study is a cornerstone for the establishment of a lean-clean methodology for earthworks on the environmental subject of sediment pollution. It is essential to highlight certain aspects of this study before proceeding to the subsequent sections. It is stressed here that any association in regards to reducing site sediment pollution in this study reflects the reduction in risk of sediment pollution, which is in-line with the aim of clean for source reduction. This is relevant because construction site sediment pollution could happen due to combination of various factors and it is challenging to pin-point it to a single factor (Belayutham et al., 2016a). Hence, this study attempts to only reduce the risk of site sediment pollution from the perspective of production factor improvement and does not require a physical sample of sediment pollution analysis. Hence, the potential relationships established in this study will only be based on the production variable. 3.2. Case study A case study method, as shown in Table 1 has been employed to examine the earthwork operation, subsequently portraying the use of lean-clean method using a real project data. In order to produce a methodology, various lean based steps are involved. Further details on the aforementioned steps are given in Section 4. Case study is a detailed investigation of a phenomenon in a real life setting (Yin, 2009). According to Christie et al. (2000), case study is advantageous when new processes are to be explored due to the rich information that can be obtained. Case study is particularly relevant for this research because it requires a focused and in-depth investigation to produce a lean-clean method that represents a real scenario. Similar method of study has been used by Cabello Eras et al. (2013) to propose CP strategies to improve the environmental performance of an earthwork project. In order to ensure the technical validity of this single case study, triangulation has been used through multiple sources of data (Yin, 2009). This case study has been conducted using several data collection methods which consist of interview, observation and site document review (site diary, daily production record, project details report). The use of different methods form a triangulation where both qualitative and quantitative methods are simultaneously applied to strengthen the academic proposition (Fellows and Liu, 2009). The case study for this research was selected based on its availability but essentially, the project must involve earthworks. The selected project is a commercial and light industrial development, located in the Raub district within the state of Pahang in Malaysia. The project has initially started off with site clearance activity that was conducted for a period of 3 months, prior to the cut and fill work that commenced from the month of June till December 2013. 4. Lean-based clean methodology The proposition for lean-clean earthworks operation is where the modification or improvement of earthwork processes using LP will potentially benefit in reducing the source of site pollution, subsequently reducing risk to the environment. Additionally, the improved processes will autonomously address the goals of LP by reduced duration that will increase customer satisfaction besides other mutual goals of both concepts such as reduced cost. This proposition has been successfully demonstrated in a study by (Belayutham et al., 2016b) whereby the improvement was observed from the perspective of administrative deficiencies.

2198

S. Belayutham et al. / Journal of Cleaner Production 142 (2017) 2195e2208

Table 1 Research work. Data collection

Data analysis

Outcome

Case study
Interview
Observation
Site document analysis

Production and environmental performance indicators

A lean based methodology to enhance the functions of clean in earthworks operation with the focus on site based water pollution.

The basic principles of LP, as introduced by Womack and Jones (1996) are used as the guiding method for this study as it provides a clear and descriptive step-by-step approach for application, which consist of: 1) Specify value; 2) Identify value stream; 3) Create flow; 4) Apply pull system and 5) Pursue perfection. Apparently, these steps are also similar to CP implementation plans prescribed by UNEP (1996), which consist of: 1) Pre-assessment; 2) Measure and identify; 3) Synthesis of information by identifying waste reduction options and 4) continuous improvement. Step 1 of lean correlates with step 1 and 2 of clean to define and measure the current situation. Then, step 3 and 4 of lean goes along with step 3 of clean to identify waste elimination opportunities and the last steps of both concept aims to pursue perfection. The similar implementation steps ensure less contradiction on implementation procedures, which will ease the use of LP procedures as the guiding protocol. Each steps in the lean-clean method contains various tools and techniques as well as data sources. Each tool does require certain data display and the methods used to attain the data are given in Table 2, together with the relevant steps. The use of those tools will be described within the body of discussion in the following section. 4.1. Step 1: Determination of VALUE Value is determined from the perspective of the customer (Womack and Jones, 1996). The first step is to recognize customer's requirements as it is the pre-requisite before further works could be done (Sayer and Williams, 2007). Assuming the customer aims to adopt lean-clean, the value should include both the goals in terms of production and the environment for an earthwork operation. The potential lean tools/practices applicable to this step are: Supplier-Input-Process-Output-Customer (SIPOC) and 5 Whys. SIPOC is a lean-based tool that is commonly used to attain the voice or requirements from customers (Sayer and Williams, 2007) while 5 Whys is a tool that has been used to identify the root cause of

problems. In this study, the SIPOC has been derived from the use of earthworks related literatures (Pain, 2014; Peurifoy and Oberlender, 2004; Martinez, 1998; Gransberg et al., 2006; Christian and Xie, 1996) and is shown in Fig. 1. The SIPOC is derived by providing answers to these questions: 1) Who are the suppliers for an earthwork operation? 2) What are the inputs required for an earthwork operation? 3) What are the processes involved? 4) What are the expected output from the operation? 5) Who are the customers of this operation? 6) What are the requirements from the customers? Answers to question 4, 5 and 6 have been searched upon first as those questions determine the core value of this operation. In general, customer is the recipient of the output from a process. In this study, the two outputs considered are production and environmental output that needs to further satisfy a set of different customers and requirements. Then, by working backwards, answers are provided for question 1, 2 and 3. From Fig. 1, the requirements could be broken down into two separate values whereby the first value would satisfy the production aspect of timely delivery, within budget and quality while the second is the environmental value of clean water and minimal emission into water bodies. As discussed in Section 1, the different values are commonly being managed in isolation through the use of production and erosion sediment control methods. In contrary, the lean-clean method proposed here intents for those distinct measures to be integrated and managed together. Further derivation is required to identify a common point of improvement for the different values identified between the customers' requirements for production and environment (sediment pollution). Point of improvement could be identified by comprehensively going through factors that could affect performances of both dimensions. In order to do so, a lean-clean technique called 5 Whys is used to derive the potential factors. 5 Whys is a lean technique used to identify the root cause of a problem (Sayer and Williams, 2007). The application of this technique has also been done with the use of literature. The procedure begins by asking the

Table 2 Lean-clean steps, tools and data sources. Lean-clean steps

Research method

Potential tools and techniques employed

Details of research method

Step 1: Value Step 2: Value Stream

Literature Review Literature Review Document analysis Observation Interviews Interview Observation Document analysis Interview Observation Interview

(Supplier-Input-Process-Output-Customer) SIPOC; 5 Whys. Value Stream Map (VSM), site productivity chart (Process Capability Measurement)

Literature review: Journal, conference, books, electronic articles, thesis. Observation: One earthwork site, 5 cycles of cut and fill. Interviews: Earthworks project personnel (Site engineers, site agent and site supervisor) with average of 15 years working experience. Document analysis: Daily site diary, daily number of trips, project document (drawing, claim payment)

Step 3: Flow

Step 4: Pull Step 5: Continuous Improvement

Root cause analysis, VSM, process capability measurement, variability Waste elimination, Pull (Just-In-Time) Standardisation, production levelling, process efficiency, cost reduction, source reduction, Kaizen

S. Belayutham et al. / Journal of Cleaner Production 142 (2017) 2195e2208

Earthwork Contractor

• •

• •

Design Engineer

•

Equipment Skilled workers

Earthwork plan Erosion and Sediment control plan

(3) Process Earthwork operation

(4) Output Production

•

(2) Input

Cut

Haul

Fill

Environment

(1) Supplier

• •

(5) Customer

2199

(6) Requirements •

Graded earth RL Platform

Land surface transformation • Excessive runoff • Erosion and sediment

• •

•

Client Next contractor

• •

• Receiving environment and its surroundings

•

Timely delivery Within budget Accepted quality Clean water Reduced emission to water body

Fig. 1. SIPOC for earthwork operation.

question ‘why’ a problem exist and the answer to the problem is written below the aforementioned problem and the same procedure will be repeated approximately five times. Similar steps will be repeated for both production (earthwork operation) and environmental aspect (water pollution). Using water pollution as an example, the derivation is given as follows: The most proximal reasons for water pollution to occur is the intertwined processes of excessive runoff, erosion and sedimentation (Chen et al., 2007). Then, the answer for why those processes occur is further searched for and could be divided into natural factors (Ismail and Yee, 2012) and man-made error (Wu et al., 2012). However, natural factors are beyond the subject of improvement under LP. Hence, only man-made factors will be the focus for further derivation. The immediate reasons for the manmade factors can further be divided into areas opened at one time (Goodemote, 2005), duration of work (Davis et al., 2003), unfavourable season (Maniquiz et al., 2009), faulty facilities (Weese, 2007) and improper practices (Yao et al., 2011). All factors could further be categorized into either pre-construction based

(distal) or construction based (proximal) factors (Belayutham et al., 2016a). However, this study concerns on the proximal factors that are within the boundary of operation. Similar method of derivation has been applied to identify factors that affect the production factor of earthwork operation. From Fig. 2, time has been identified as the point of similarity for improvement since both aspects of production and environment are being affected by time. Therefore, efforts to improve the time variable could first improve the production performance by completing the operation ahead of schedule or at least on-time and secondly, time reduction could reduce the occurrence of rainfall, subsequently reducing the risk of excessive runoff and erosion, ultimately minimising the risk of water pollution. This step has provided a theoretical based integration in identifying the similar factor for simultaneous improvements of the production and environmental aspect of earthwork operation. Hence, the rest of the lean-clean steps will be scoped to address the time factor.

Fig. 2. Point of similarity for earthwork production and environmental improvement.

2200

S. Belayutham et al. / Journal of Cleaner Production 142 (2017) 2195e2208

4.2. Step 2: Identification and mapping of the VALUE STREAM This section aims to identify and document the current process of earthworks operation by recognizing its work progression from a short and long duration scale. Short time period allows the identification of inefficiencies in processes while long time period could provide a perspective in recognizing variability of the production. This mostly measurement related step will establish indicators to measure the current status of earthwork from the production and environmental aspect. Detailed discussion on the measurements and the portrayal of an earthwork operation will be given in the following section, which is organized into production and environmental measures. The Lean-Clean tools/practices adopted for this step are as follows: Value Stream Map (VSM) and site productivity chart (process capability measurement). 4.2.1. Production measure In this study, the production measure is proposed to be identified at two levels: 1) Micro level: Detailed perspective of the production process by mapping the VSM that identifies operation waste and 2) Macro level: A larger perspective that shows the operation's monthly output through the use of bar chart that displays output variations. It is essential to highlight that days with only soil material as an output will be considered for this study. The intention is to provide a comparable platform to identify waste and inefficiencies in the process. Soil accounts for 80.1% of total material from the site under study that includes also rock at 11.9% and hard material at 8%. At this stage, taking into consideration all the other materials will only complicate this initial attempt of portraying the use of lean-clean approach. This is due to the potential outlying reason for inefficiencies that can be related to characteristics of the material, which is beyond the control of the production team. However, the mentioned scope of work and limitation could be a point for future studies. Micro level (VSM) VSM is a graphical representation of the flow of processes, information and material in a system that delivers output. Mapping allows the sources of waste in a system to be identified and eliminated (Rother and Shook, 2009). In this study, VSM is used to portray current processes involved in the earthwork operation that comprised of three main processes, i.e. cut, haul and fill (Martinez,

Table 3 Measured indicators. Indicators

Measurement

Indicators

Measurement

Start day Finish day Start time Finish time Idle time

Month/day Month/day Hour/mins./sec. Hour/mins./sec. Hour/mins./sec.

Distance travelled No. of workers No. of equipment No. of trips

m No. No. No.

1998). Data for the purpose of producing this VSM has been collected from an on-site observation as well as discussion with the project personnel involved in the project. In order to further derive the performance metrics of the operation, productivity scale of the operation should first be defined. Generally, earthwork productivity could be measured with volume of earth per unit of time (m3/t) which can be derived from the number of unloading or trips by the trucks in a day. Capacity of the trucks used in this project is 6 m3. Hence, 1 truck unload/trip ¼ 6 m3 of soil. The truck load can be converted to quantity of soil where the flowing unit is determined as m3 of earth. In order to portray the details of earthwork in a VSM, data for the required indicators are shown in Table 3 (NZ Qualification Authority, 2015): From the indicators, production factors could be processed and is given as follows:
Cycle time (min./sec.) ¼ The definition of cycle time by Hopp and Spearman (2011) is used where cycle time is the average time for a job to go through a production line.
Productive time (PT) ¼ Duration when earth is being worked on.
Non-productive time (NPT) ¼ Duration when no movement of earth is observed.
Output per day (m3) ¼ (No. of trips x 6). Other variables which could also provide measures for the work progression include downtime of equipment and frequency of downtime. This information was not obtained from observation but has been provided by the site engineer, which is further verified with the site diary. For the case studied, the current VSM based on the aforementioned process and indicators is shown in Fig. 3. The VSM is drawn

Fig. 3. Current map for earthwork operation.

S. Belayutham et al. / Journal of Cleaner Production 142 (2017) 2195e2208

based on five cycles of observation with the same haulage distance. After discussion with the site engineer, it is agreed that a single cycle of work would best be represented by 10 unloads/trips to portray the earth movement of the related equipment. Therefore, 1 cycle ¼ 10 unloads/trips of earth. It is essential to highlight that the time value provided in the VSM is represented as the average time. VSM tries to identify opportunities for improvement rather than to produce very accurate measures of performance, thus, average time is considered reasonable (Rother and Shook, 2009; Rosenbaum et al., 2014). VSM has provided measures for the operation at the process level. However, it is difficult to distinguish the variability of output achievement for a longer period of time. Therefore, a macro perspective on the performance of the earthwork operation is proposed and the information could be obtained from the archival data (daily productivity that is represented by number of trips by trucks in a day) of the project. This macro perspective could not directly pin-point the operational waste but allows the portrayal of output variability where further enquiries could be done to identify reasons for the variability and potential waste where variability is significant. For this study, the information obtained is for a period of 7 months where all data is represented in a monthly basis (Refer to Fig. 4). The monthly output portrays a large variability in output between the different months. This macro perspective could complement the limitations of the conventional VSM by portraying the work performance over a longer period of time, consequently enabling the identification of the output patterns. The performance pattern allows the assessment of variability in order to identify the smoothness or unevenness of the output. According to Deif (2012), previous studies have focused much on waste elimination while variability elimination has been side-lined. From a lean perspective,

2201

it is preferable for output to be smooth rather than to have large variances because variability could signal subsequent problems such as congestion and longer lead time (Deif, 2012). In this study, the bar chart of output in Fig. 4 is supplemented with the Coefficient of Variation (CoV) analysis. Thomas and Zavrski (1999) has used CoV to measure variability where higher value of CoV indicates higher variability in the system. Similarly, Deif (2012) has also used CoV to capture time and flow variances in his study while Shehata and El-Gohary (2011) concluded that the criteria to improve project performance is through reducing variability in output. In this study, the highest CoV or variability in daily output is for the month of July. Hence, further analysis is conducted to dig deeper into the output for the month of July in order to understand the causes of the variability. The summary of output for the month of July is given in Fig. 5. Further details of the production level will be discussed in the following section. 4.2.2. Environmental measure Rainfall is one of the most important agent for soil erosion that could potentially increase the risk of sediment pollution (Lee and Lin, 2015). From the perspective of environment (sediment pollution), time is relevant when being calculated alongside rainfall occurrence within the period of time, in order to produce the value of rainfall erosivity. Ideally, improved earthwork processes could lead to shortened earthwork operation time, consequently lowering the risk of sediment pollution. The claim could be proven using the Universal Soil Loss Equation (USLE). USLE is an erosion model designed to predict the long-term average annual soil loss from specific areas (Wischmeier and Smith, 1978). This equation has been tested and validated throughout the years since its establishment (Teh, 2011). The USLE equation involves six

Monthly No. of Trips 5382

6000

5289

Month June July August September October November December

3514

4000 3000

1923

1435

2000

948

1000

685

0 Jun

Jul

Aug

Sep

Oct

Nov

CoV 0.01633 0.18661 0.0946 0.11836 0.08154 0.17093 0.17162

Dec

Month

Fig. 4. Monthly output.

Daily Truck Output 350

324

298

300 Total Trips

No. of trips

5000

250 226 232

268

264 257

292

276 278 225

208

241

233

225

150

287

264 206 209

189

200

266

114

100 50 0 1

2

3

4

5

6

7

8

9

10 11 12 13 14 15 16 17 18 19 20 21 22 Days

Fig. 5. Performance for the month of July.

2202

S. Belayutham et al. / Journal of Cleaner Production 142 (2017) 2195e2208

variables: 1) rainfall and runoff erosivity (R); 2) soil erodibility (K); 3) slope length (L); 4) slope steepness (S); 5) cover/management practices (C) and 6) supporting conservation practices (P), with the equation given as follows: A¼RKLSCP Where, A is the computed spatial and temporal average soil loss per unit of area. This study intent to portray the relationship between reduced production time and sediment pollution. According to King and Holder (1977) and Balousek (2000), the major variable that influences the prediction of soil loss over a short time period such as construction is the ‘R’ factor. Hence, only the ‘R’ factor would be of concern as it relates to the variable of interest, which is time. As a guide, lower ‘R’ factors will reduce the probability of storm-induced erosion (Isikwue et al., 2015). For this study, the steps proposed by King and Holder (1977) in evaluating soil loss for a short time period has been adapted. Step 1: To identify area and time period of study where the USLE factors can be assumed to be constant. Step 2: To calculate the adjusted ‘R’ factor based on the Erosivity Index distribution curve for the period of study. The unit of ‘R’ value in this study is MJ.mm/ha.yr. Step 3: To evaluate the different values of the ‘R’ factor with respect to the different time period considered. Relate the value to soil loss. Balousek (2000) have also adapted the USLE equation in their study for predicting soil loss from construction sites. Their contribution in adapting the USLE into shorter period highlighted the importance of the time element that indicates the criticality of erosion for certain period of a year. The current earthwork project acquired a duration of seven months for completion, from June to December. Hence, the calculated rainfall erosivity ‘R’ for the given operation period is 10,065 MJ mm/ha.yr. Detailed calculation of the ‘R’ factor is given in the Appendix.

4.3. Step 3: Flow creation through waste elimination This current step functions to identify reasons for the waste found in VSM and variability identified in the monthly output chart. In the language of lean, this step identifies the inefficiencies that deter smooth flow of the earthwork operation. Improvements to the identified reasons could further improve the production factor of time, consequently also the ‘R’ factor of the environmental dimension. The lean-clean tools adopted in this step are as follows: root cause analysis; VSM; process capability measurement and variability.

4.3.1. Waste recognition from VSM In reference to the current VSM given in Fig. 3, the process flow shows both the productive (cycle) and non-productive (idle) times of the earthwork operation. The percentage of non-productive (idle) time is 31.5% of the total lead time of 124 min. A large portion of the idle time (15 min) is found at the fill area where the dumped soil is not being worked on till it reaches 10 loads. Even though the idle time did not cause congestion to other work sections, the two idling machineries (dozer and compactor) represent waste of resources. Another apparent waste can be seen at the cutting area where it shows the idle time of excavator when soil is not being worked on. Based on a discussion with the site engineer of the project under study, the resources provided for loading and hauling are two excavators with five trucks, where the truck will move from cut (C) to fill (F) area. The site engineer also mentioned that 1 excavator will service between 2 and 3 trucks for a round trip. Using the cycle time given in the VSM, it is observed that excavator loading time is 2 min per trip while the haulage time for truck is 6 min per trip. The movement and position of trucks is shown in Fig. 6. When the excavator services 3 trucks in a round trip, the excavator will be idle for at least 2 min (dashed line represents non-working time for the excavator) while waiting for the first truck to return (Refer to Fig. 6, Excavator 1). At the same time, Excavator 2 would be servicing 2 trucks and the waiting time by the other excavator for the return of truck 1 is 4 min (dashed line represents non-working time for the excavator) (Refer to Fig. 6, Excavator 2). Theoretically, calculation for the required number of trucks with one excavator is given in the following equation by Gransberg et al. (2006), which was recognized earlier by Peurifoy and Oberlender (2004). It is essential to highlight that the equation uses average production rate which is deterministic in nature. Number of required trucks ¼ truck (loading þ going þ return þ dumping) time / loading time For the current site, the total loading, hauling and dumping time is 8 min while the loading time is 2 min (Refer to VSM in Fig. 3). Hence, a single excavator is expected to serve 4 excavators where two excavators could serve up to 8 trucks. It is apparent that the current project is running low on trucks, causing inefficient use of the excavators. Additionally, a smaller portion of the total idle time can be seen at the cut and fill area with 5 min idle time observed for each cycle as the truck prepares to move and navigate to or from the area. It is essential to reiterate here that the time measurement is given as average time and is deterministic in nature. There is no doubt that there are more precise modelling options available such as computer modelling that could provide a stochastic value (Poshdar et al., 2014). However, the use of stochastic value is not necessary in relation to the purpose of this analysis, which is to identify improvement opportunities at the current site, which are not visible using traditional management methods.

Fig. 6. Movement of truck.

S. Belayutham et al. / Journal of Cleaner Production 142 (2017) 2195e2208

4.3.2. Variability based on production output From the macro perspective, the bar chart given in Fig. 5 shows a relatively uneven work output between days in July. According to Thomas and Zavrski (1999), variability in daily output has a strong correlation with project performance. Hence, the daily output is further investigated by representing the daily output of trucks (Truck A, B, C, D, E) against the working days in Table 4. The information has been gathered from the particular site's daily record that consist of number of trips by trucks. Furthermore, CoV value is provided to show the variance between total number of trips daily, which is given as 18.66%. The highest output is given as 324 trips on day 21 while the lowest output is 114 on day 16. A correlation between variance and daily output performance is identified among the days with high variance (shaded rows in Table 4) with its respective number of trips by using Pearson's r correlation. Result of the correlation is given as 0.8146, which indicates a strong negative correlation, whereby increase in variability will affect daily performance by the decrease in daily total trips. Due to the negative effect of high variability, the site personnel that consist of site engineer and site supervisor from the contractor's organization and site agent who represents the client were first queried on possible reasons for the monthly output unevenness. The interview with them started off with question why large variability of output is observed between different days. From the input, the major reason is related to machine breakdown, followed by rainfall and skill of operators. Besides the deficiencies, it has also been highlighted that the distance may vary between days and this could have contributed to the different outputs, apart from the discovery of rock and other hard materials. Advancing from that, it is also observed that the number of trips differ quite significantly from one truck to another. Refer to Table 4 on the computed CoV based on the daily output of trucks. From the table, Truck B has the highest output while Truck E has the lowest output. However, due to slight changes in distance between days, the comparison between days might not reflect an ‘apple to an apple’ comparison. According to Thomas and Zavrski (1999), daily variability can be used to set apart good and bad performing projects. Hence, detailed identification of inefficiencies is done based

2203

on daily output between trucks as this will provide a common ground for comparison due to similar haulage distance in a day. From Tables 4 and 8 out of 22 days provide CoV value greater than 10%. Due to that, the site based respondents have been queried on the reasons for the difference in output between trucks of the same distance. The respondents were asked the following question: From the data, particularly on days with high CoV (shaded in Table 4), we could see that the daily total trips do vary between trucks. Can you please provide factors that could have caused such variance between the trucks? From the interview, the related factors that could have caused the unevenness are given as follows:
Machine breakdown, especially when there are no backup resources such as the breakdown of Truck A on Day 15 which has contributed to the highest CoV of the month.
Skill and experience of the equipment operator, where the difference between experienced and less experienced ones could result in shortages of at least 12 m3/day or 2 trips of tipper, as shown in the lowest CoV day, which is day 2. The differences in skill provides a non-standardised work execution that causes disruption in the flow of work that can contribute to disturbance in haulage time, further prompting unnecessary queuing and unevenness in output between trucks.
At the cut section, the position and turning point/swivel degree of the excavator creates differences in time and efficiency. Smaller swivel point is much efficient than large swivel points. Hence, if the truck operator could position the truck to reduce the motion of excavator, the cycle time could be shortened.
At the fill section, cycle time increases when tipper unloads soil far from the dozer. Common improper practices can also be found with compactors where vibrators were not activated in attempt to reduce cost. Non-vibrated compactor could cause a longer cycle time besides further damages such as failing compaction test that leads to unnecessary halt of the operation. Those aforementioned reasons also coincide with certain causes of variability in manufacturing such as different processing time and unavailability of machine, as mentioned by Deif (2012).

Table 4 CoV for daily truck output. Working days

No. of trips Truck A

Daily total trips Truck B

Truck C

Truck D

Daily CoV between trucks

Truck E

1

47

50

42

50

37

226

0.12456

2 3 4 5 6

47 41 60 57 54

47 43 54 51 70

46 47 49 48 69

47 38 48 49 52

45 39 53 52 53

232 208 264 257 298

0.01928 0.086 0.09024 0.06823 0.15221

7 8 9

60 53 47

55 69 50

51 68 41

50 51 50

52 51 37

268 292 225

0.07532 0.15861 0.12862

10 11 12

62 62 46

57 57 50

53 53 42

51 52 50

53 54 37

276 278 225

0.07938 0.07261 0.12373

13 14 15 16

37 48 0 41

39 47 28 41

43 46 28 41

34 47 29 59

36 45 29 59

189 233 114 241

0.09049 0.02447 0.55945 0.20454

17 18 19

55 64 32

55 59 55

53 56 54

52 53 33

51 55 32

266 287 206

0.03363 0.07453 0.29498

20 21 22 Total Trips CoV for daily trips

42 71 59 1021

43 62 53 1076

47 62 49 1032

38 64 49 993

39 65 54 973

209 324 264

0.08526 0.05712 0.07855

0.18661

2204

S. Belayutham et al. / Journal of Cleaner Production 142 (2017) 2195e2208

4.4. Step 4: respond to customer Pull

Table 6 Cost of equipment hire.

Generally, this section discusses on the solutions for the inefficiencies and waste identified at the micro level which used the VSM. The tools that will be adopted in this step are waste removal, Just in Time (JIT) and pull system. From the micro perspective of VSM, waste could be identified from the current processes (Fig. 3) where activities are linked to each other using the traditional push system. Earth is loaded by the excavator into the truck, which is then passed on to the fill section. Mismatch happens when push is being applied without matching the availability of truck that results in idle time for the excavator, as shown in Fig. 6. The solutions proposed to address the inefficiencies is either by 1) reducing the number of excavator to 1 with trucks reduced to 4 or 2) retain both excavators but to add 3 trucks into the system. This is in order to eliminate the idle time of the excavator that waits for the trucks to return. The truck requirement is calculated using the deterministic equation given in step 3, with the use of the measured current loading and hauling time taken from the VSM. The ideal set of trucks to create flow by eliminating the excavator idle time is 1 excavator ¼ 4 trucks. In order to improve the time factor, which will then benefit the production and environmental measure, option 2 is the preferred choice as option 1 will cause delay on the time factor whilst the intended aim is to expedite the completion of the operation. Hence, the current study would require the addition of 3 trucks to ensure no waste of waiting in the operation. The calculated outcome of this chosen option is given in Table 5. It is shown that with the addition of 3 trucks, the soil hauling output can be increased by 60% from the original 5 trucks. The number of trips required for this project is approximately 19,176 trips, equivalent to approximately 115,056 m3 of soil. With the addition of 3 trucks, the number of trips could be achieved as early as the 7th operation day in November. Hence, work could be completed by the 7th operation day in November. The reduction in time has a positive consequence on the risk of sediment pollution where the calculated ‘R’ factor is reduced to 7672.5 MJ mm/ha.yr. Nonetheless, the issue of cost might be of concern as increase in resources could cause the increase in cost. Hence, a cost analysis based on direct cost of equipment hire is conducted to ensure the viability of the improvement strategy. Table 6 shows the total cost of equipment hire for the present situation and also the improved situation with added trucks. The original duration of the project is 75 days and the increase in truck could shorten the operation to 52 days. The daily rate for those equipment was given by the site engineer. The total cost of adding trucks is still lower than the total cost with no added truck but longer operation time. Hence, the proposed strategy benefits the production (time and cost) and environmental ‘R’ factor for water pollution.

Equipment

Truck Excavator Compactor Dozer

Daily rate (RM)

500 550 600 650

Original no. of trucks

With added trucks

No.

Hire cost

No.

187500 82500 45000 48750 363,750

8 52 2 1 1 Total cost

Days

5 75 2 1 1 Total cost

Days

Hire cost 208000 57200 31200 33800 330,200

4.5. Step 5: CONTINUOUS IMPROVEMENT to pursue perfection The macro level inefficiencies perceived from the variability in output will be addressed here using lean tools and principles such as standardisation, production levelling, process efficiency, cost reduction, source reduction and kaizen. Reasons for the variation between trucks have been determined in Step 3 where the respondents have generally related the variations back to machine breakdown and skill of operators. In order to pursue perfection, an ideal situation is desired. For an ideal situation, the concept of maximum production should be in use by eliminating all waste in the operational processes. To improve current processes, the illpractices identified in Step 3 should be eliminated. Firstly, in order to address machine breakdowns, proper maintenance schedule should be in place. Hence, it is proposed for the organization to plan a preventive maintenance program alongside the construction schedule in order to enable maintenance works to be conducted. LP has heavily emphasized the importance of equipment maintenance with a simple method called preventive maintenance (Ohno, 1988). Conversation with the site engineer has revealed that they do not have a fixed maintenance schedule for each operator where the common practice is to repair the equipment once it is dysfunctional. The second strategy for improvement is to fully utilize the resources by eliminating idle time. Therefore, it is essential for operators to be educated on the right and most optimized way of handling the equipment. In order to do that, a baseline productivity should be attained, where it represents the best performance with minimal to almost no waste or disruption (Abdel-Razek et al., 2007). The steps to achieve the baseline productivity should be standardized among all operators. Following that, unnecessary truck idle time could be eliminated as operators have been taught on the proper way of positioning the truck for ease of movement at the cut and fill area. For illustration purpose, supposedly, the maximum output truck would have done the work with minimum to almost no idle time. Hence, it is suggested that other truck operators could also perform at the maximum level when work is standardized and idle time eliminated. Hence, the output of trucks could be levelled up to the highest production of the day. Ideally, if potential output levelling can be achieved following each month's maximum production with no machine breakdown, the operation can be completed by the 5th

Table 5 Outcome with additional trucks. Month

Current no. of trips

Cumulative no. of trips

June 1435 1435 July 5382 6817 August 948 7765 September 685 8450 October 1923 10373 November 3514 13887 December 5289 19176 Original ‘R’ factor: (1.000e0.39)*16,500 ¼ 10,065 MJ mm/ha.yr. ‘R’ factor with 3 additional trucks: (0.855e0.39)*16,500 ¼ 7672.5 MJ mm/ha.yr.

With 3 additional trucks (1.6 of current trip)

Cumulative no. of trip with added truck

2296 8611 1517 1096 3077 5622 8462

2296 10907 12424 13520 16597 19545 7th day November

S. Belayutham et al. / Journal of Cleaner Production 142 (2017) 2195e2208

day in December. Even though it does not seem to be much, relatively it could reduce up to 12 operation days. 4.5.1. Overview on the proposed improvement strategies In order to put all the improvement options into perspective, an Ideal Map is presented in Fig. 7 with the expected productivity and environmental performance improvements (shown in Table 7). The ideal map is drawn over the current map to envisage the elements to be removed and improved. All the expected to be eliminated (shown as cross in the ideal map) elements is envisioned as a result of having proper preventive maintenance (elimination of downtime and frequency of downtime), standardized work (elimination of the 5 min between task), using pull system to replace push system and adding 5 to 8 number of trucks. On the other hand, improvement (shown as oval in ideal map) is expected for the productive times of each process when work standardization is being implemented. The plan for improvement is preferred by first removing time and resource waste through the addition of trucks. This is followed by stabilisation of production as discussed in the current step by addressing the issue of maintenance and operator efficiency. The first solution could be applied to solve issues during construction itself as VSM provides an instant recognition of waste. The second solution which is based on variability of output might

require a longer period of observation as patterns of output needs to be discovered. Overall, proposed strategies should be implemented progressively towards reaching the ideal state where the concept of Kaizen is to be practiced in aim to continuously improve the value stream towards pursuing perfection. For this study, the combined approach could result in the project completing by the 5th day of operation in October and can be viewed in Column 6 and 7 of Table 7. This time reduction could consequently provide an ‘R’ factor of 5857.5 MJ mm/ha.yr with 42% positive improvement. The combined approach also resulted in changes to cost requirement and this is represented in Table 8 where cost for the additional hire of equipment is given. Further reduction in duration due to the stabilisation effort could reduce the duration to 43 days that relatively results in lower cost compared to the original setting and the setting with only improvements of adding trucks (Refer to Table 7). The proposed integration has received the acknowledgement from the site personnel as they found this integration interesting because they have never thought of managing both aspects of production and environment concurrently. They also mentioned that the strategies are certainly beneficial as it is proven to improve the production factors of time and cost, besides the reduced risk of sediment pollution. Furthermore, they also commended on the ease of

Fig. 7. Ideal map. Table 7 Proposed Outcome.

2205

2206

S. Belayutham et al. / Journal of Cleaner Production 142 (2017) 2195e2208

Table 8 Direct cost of Equipment Hire. Equipment

Truck Excavator Compactor Dozer

Daily rate (RM)

500 550 600 650

Before improvement

After improvement

No.

Hire cost

No.

187500 82500 45000 48750 363,750

8 43 2 1 1 Total cost

Days

5 75 2 1 1 Total cost

Days

Hire cost 172000 47300 25800 27950 273,050

applying the method, especially on improving the earthworks operation. The operation which involves less trades is definitely a plus point as it could facilitate the motion of changing the current management system. 5. Results and discussion The proposed lean-clean methodology was grounded to the five core principles established by Womack and Jones (1996) with tools and techniques adopted from reputable lean experts such as Ohno (1988), Monden (1993) and Shah and Ward (2003). The methodology has resulted in improvements for both the production (time and cost) and environmental (risk of site sediment pollution) measures. Positive improvements have been observed in terms of time and cost reduction by 42.7% and 24.9% respectively. The environmental factor, Rainfall Erosivity is reduced by 41.8%, consequently reducing the risk of soil erosion and sediment production. Pampanelli et al. (2014) who have developed similar concept model, Lean & Green that was based on the principles of lean with tools suited to their study was able to reduce 30%e50% on average the use of resources and 5%e10% reduced total cost of mass and energy flows in a production cell. The proposed lean-clean methodology in this study that encourages simultaneous application of lean and clean also aligns with the study by Galeazzo et al. (2014), who found that pollutionprevention initiatives may benefit when the mutual interdependence capability between lean and green is capitalized to address both dimensions of operation and environment. In general, the present methodology and result echoes findings from previous studies, thus suggesting that lean could enable clean. This study has the potential to be replicated to other earthworks sites that seek to improve their operation. By using the proposed methodology, observation in as short as one week could provide the variability and daily output data for fast improvements. However, for different operations or environmental concern, the proposed methodology could also be used provided that the ‘value’ must be determined at the very initial stage. The ‘value’ should be mutual to address both the production and environmental dimensions. In the case of this study, the ‘value’ is time. Furthermore, the environmental measure determined in Step 2 should also correspond with the selected environmental concern. 6. Conclusion This study has developed a lean-clean methodology for an earthworks operation in regards to the environmental subject of site sediment pollution. The application of lean has shown great potential to benefit both the production (time and cost) and environmental (sediment pollution) performance of an earthwork operation, subsequently enabling clean. However, the various steps involved in the method could contain certain limitations with some worth mentioning. The calculation for number of trucks has been given in a deterministic manner, whereas earthworks itself is dynamic with various interaction between factors. This is well understood by the authors but to delve into the subject of

deterministic and stochastic will divert the original intent of this study, which is to explore the subject matter and propose opportunities to identify and act on production and environmental waste in construction. Nonetheless, this limitation creates a window of opportunity for future studies by taking stochastic variables and computer simulation into consideration, subsequently moving this study towards confirmatory research. In addition, the proposed solution is specifically to address issues in the case under study. Hence, solutions may vary depending on different cases as every construction project is unique. Hence, more studies should be done to test the applicability of this method in various industry settings. This study has filled the current gap of knowledge which lacks detailed methodology to support the implementation of clean. The detailed step-by-step portrayal of lean in the earthworks operation has provided a transparent demonstration of how both variables of production and environment could seamlessly be improved. This is also a pioneer work that associates the environmental variable of sediment pollution and production factor as the previous study on this matter is very limited. Thus, an environmental measure for evaluating the risk of site sediment pollution in correspondence to the production factor has also been established. The current association has also contributed to the body of knowledge of the individual concepts of lean and clean. For lean, the integration has advanced from its original production-based functionalities to include the environment. For clean, the integration has remedied some grey areas within the concept, especially on process modification. In the practical world, the lean-clean method could assist practitioners in viewing the importance of production and environmental measures simultaneously, so that imbalance treatments of the two different dimensions could be reduced. Furthermore, the techniques introduced could assist practitioners to identify their inefficiencies in a more systematic manner whether it is for a short term solution by identifying waste using VSM or long term-based by observing the variability in output. This simple-step-by-step methodology requires only data that are easily available at the site. It does not require advanced technologies or knowledge that could be an Achilles’ heel among practitioners. Overall, this development could benefit the construction industry in specific and people at the receiving end in general. Acknowledgement The authors would also like to thank the industry respondents for their valuable insight and expertise that have greatly benefitted the study. Appendix Example of computation steps for the Rainfall Erosivity ‘R’ factor is given as follows: Step 1: The area studied is located in the state of Pahang. The Rainfall Erosivity value for this site is between 16, 000 to 17, 000 MJ mm/ha.yr. This figure could be found in the Pahang State Isohyet Map, given in DID (2010). Based on the site location, the average rainfall erosivity value of 16, 500 MJ mm/ha.yr is selected to represent the annual ‘R’ factor. Step 2: The monthly modification factor (% rain monthly) to calculate the adjusted ‘R’ factor is given in Table A1. This is required as the ‘R’ given in Step 1 is not monthly based. The monthly modification factor can be referred to in DID (2010). The cumulative value of the % rain was calculated from the month of January but only values for the affected month is shown here. The short term based ‘R’ factor can be calculated by multiplying the cumulative rainfall for the affected month with the annual ‘R’ factor selected in Step 1.

S. Belayutham et al. / Journal of Cleaner Production 142 (2017) 2195e2208

2207

Table A1 ‘R’ Factor Computation. Month

May

% Rain monthly Cumulated % rain ‘R’ for 7 months (Before lean-clean improvement)

0.065 0.060 0.060 0.060 0.39 0.455 0.515 0.575 (1.000e0.425)*16,500 ¼ 10,065 MJ mm/ha.yr.

Jun

Step 3: The lower value of the modified ‘R’ factor for after leanclean improvement suggests a lower soil erosion rate due to rainfall. Hence, the reduction in time land being left opened could reduce the risk of sediment pollution. References Abdel-Razek, H.A., Hany, A.M., Mohammed, A., 2007. Labour productivity: benchmarking and variability in Egyptian projects. Int. J. Proj. Manag. 25, 189e197. Baines, T., Brown, S., Benedettini, O., Ball, P., 2012. Examining green production and its role within the competitive strategy of manufacturers. J. Ind. Eng. Manag. 5 (1), 53e87. Balousek, J.D., 2000. Predicting Erosion Rates on Construction Sites Using the Universal Soil Loss Equation in Dane County, Wisconsin. National Conference on Tool for Urban Water Resource Management and Protection, Chicago, IL.  Delgado, J.A., Acosta, F.J., 2013. ndez-Sa nchez, G., Berzosa, A., Barandica, J.M., Ferna Applying life cycle thinking to reduce greenhouse gas emissions from road projects. J. Clean. Prod. 57, 79e91. lez, V.A., 2015. A lean approach to manage production and Belayutham, S., Gonza environmental performance of earthwork operation. In: Proc. 23rd Ann. Conf. Of the Int'l. Group for Lean Construction. Perth, Australia, July 29-31, pp. 743e752. Belayutham, S., Gonz alez, V.A., Yiu, T.W., 2016a. The dynamics of proximal and distal factors in construction site water pollution. J. Clean. Prod. 113, 54e65. lez, V.A., Yiu, T.W., 2016b. Cleane-lean administrative proBelayutham, S., Gonza cesses: a case study on sediment pollution during construction. J. Clean. Prod. 126, 134e147. Bergmiller, G.G., Mcwright, P.R., 2009. Lean manufacturers' transcendence to green manufacturing. In: Industrial Engineering Research Conference, Miami, Fl. Brown, W.E., Caraco, D.S., 1997. Muddy water in, muddy water out? A critique of erosion and sediment control plans. Watershed Prot. Tech. 2 (3), 393e403. rrez, A.S., Capote, D.H., Hens, L., Vandecasteele, C., 2013. Cabello Eras, J.J., Gutie Improving the environmental performance of an earthwork project using cleaner production strategies. J. Clean. Prod. 47, 368e376. Capony, A., Lorino, T., Muresan, B., Baudru, Y., Dauvergne, M., Dunand, M., Colin, D., Jullien, A., 2012. Assessing the productivity and the environmental impacts of earthwork machines: a case study for GPS-instrumented excavator. Proced. Soc. Behav. Sci. 48, 256e265. Chen, T., Cui, P., Chen, X., 2007. Prediction of Soil Erosion on Different Underlaying Surface in Construction Period of Xichang to Panzhihua Expressway, vol. 12. Wuhan University Journal of Natural Sciences, pp. 699e704. Christian, J., Xie, T.X., 1996. Improving earthmoving estimating by more realistic knowledge. Can. J. Civ. Eng. 23, 250e259. Christie, M.J., Rowe, P.A., Perry, C., Chamard, J., 2000. Implementation of realism in case study research methodology. In: Proceedings of Entrepreneurial SMESengines for Growth in the Millennium: International Council for Small Business World Conference. Brisbane: Queensland, June 7-10, 2000. Cleef, T., 2014. Exploratory Data Analysis in Business and Economics: an Introduction Using Spss, Stata, and Excel. Springer International Publishing, Switzerland, pp. 1e22. Cobra, R.L.R.B., Guardia, M., Queiroz, G.A., Oliveira, J.A., Ometto, A.R., Esposto, K.F., 2015. Waste” as the common “gene” connecting cleaner production and lean manufacturing: a proposition of a hybrid definition. Environ. Qual. Manag. 25 (1), 25e40. Davis, C.R., Johnson, P.A., Miller, A.C., 2003. Selection of erosion control measures for highway construction. In: World Water and Environmental Resources Congress, Pennsylvania: USA, June 23-26, pp. 262e271. Dawood, N., Chavada, R., Benghi, C., Sanches, R., 2010. Interactive visual lean system for resources planning of earthwork operations. In: 18th Annual Conference of the International Group for Lean Construction, Haifa, Israel. IGLC. Degani, C.M., Cardoso, F.F., 2002. Environmental performance and lean construction concepts: can we talk about a 'Clean Construction'?. In: Paper Presented at the 10th Annual Conference of International Group for Lean Construction, August 68, Gramado, Brazil. Deif, A., 2012. Assessing lean systems using variability mapping. Procedia CIRP 3, 2e7. DID (Department of Irrigation and Drainage), 2010. Guideline for Erosion and Sediment Control in Malaysia. Department of Irrigation and Drainage Malaysia, Kuala Lumpur: Malaysia, pp. 61e87. Fellows, R.F., Liu, A.M.M., 2009. Research Methods for Construction. Wiley. Fidler, K., Betts, S., 2008. Lean Earthworks. LCI Institute UK Summit, England. Fu, J., 2013. Logistics of Earthmoving Operations-simulation and Optimization. KTH Royal Institute of Technology. Licentiate Thesis.

Jul

Aug

Sep

Oct

Nov

Dec

0.075 0.65

0.095 0.745

0.110 0.855

0.145 1.000

Galeazzo, A., Furlan, A., Vinelli, A., 2014. Lean and green in action: interdependencies and performance of pollution prevention projects. J. Clean. Prod. 85, 191e200. Geiser, K., January-June 2001. 2001. Cleaner Production perspectives 2: integrating CP into sustainability strategies. UNEP Ind. Environ. 33e36. Giacchetti, M.C.M., Aguiar, A.O., 2015. Cleaner production: a bibliometric study in scopus data base. In: 5th International Workshop-Advance in Cleaner Production. Sao Paulo: Brazil. 20e22 May 2015. lez, V., Poshdar, M., 2013. Improving construction environGolzarpoor, H., Gonza mental metrics through integration of discrete event simulation and life cycle al, Canada, pp. 130e139. analysis. In: 30th ISARC, Montre lez, V., Echaveguren, T., 2012. Exploring the environmental modeling of road Gonza construction operations using discrete-event simulation. Autom. Constr. 24, 100e110. Goodemote, G.J., 2005. Keeping it real. Public Works 136, 56e58. Govenor, T., 2008. Green construction. Ohio BWC’s Division of Safety & Hygiene. https://www.bwc.ohio.gov/downloads/blankpdf/SafetyTalk-Greenconstruction. pdf (Accessed 4 January 2016). Gransberg, D., Popescu, C., Ryan, R., 2006. Construction Equipment Management for Engineers, Estimators, and Owners. Taylor and Francis Group, Boca Raton, FL. Hamner, B., 1996. What is the relationship between cleaner production, pollution prevention, waste minimization and ISO 14000?. In: The 1st Asian Conference on Cleaner Production in the Chemical Industry, December 9e10, Taipei, Taiwan. Harbor, J., 1999. Engineering geomorphology at the cutting edge of land disturbance: erosion and sediment control on construction sites. Geomorphology 31, 247e263. Hopp, W.J., Spearman, M.L., 2011. Factory Physics, third ed. Waveland Press, USA. Huisingh, D., Bass, L.W., 1991. Cleaner production: the most effective approach to achieving improved water quality. Eur. Water Pollut. Control 1, 24. Isikwue, M.O., Ocheme, J.E., Aho, M.I., 2015. Evaluation of rainfall erosivity index for Abuja Lombardi method. Niger. J. Technol. 34 (1), 55e63. Ismail, S.I.H., Yee, H.M., 2012. MUSLE evaluation of soil loss on a construction site by using gauging weirs. Adv. Mater. Res. 446e449, 2718e2721. Jaeger, R.G., Halliday, T.R., 1998. On confirmatory versus exploratory research. Herpetologica 54, S64eS66. Kaiser, J., Zikas, T., 2009. Lean Management in Road and Underground Construction. Bauportal [Online]. Available At. http://Www.Building-ConstructionMachinery.Net/Shop/Topics/Earthworks-Compacting/Topic/Lean-ManagementIn-Road-And-Underground-Construction/ (Accessed 3 January 2015). Kang, S., Seo, J., Baik, K., 2009. 3D-GIS based earthwork planning system for productivity improvement. Constr. Res. Congr. 2009, 151e160. Kemppainen, J., Makinen, J., Seppanen, O., Kankainen, J., 2004. Lean construction principles in infrastructure construction. In: 12th Annual International Group for Lean Construction Conference, Helsingør, Denmark. IGLC. King, A.D., Holder, T.J., 1977. Preliminary Guidance for Estimating Erosion on Areas Disturbed by Surface Mining Activities in the Interior Western United States. U.S. EPA. Kirchbach, K., Koskela, L., Gehbauer, F., 2014. Digital Kanban for Earthwork Site Management. In: 22nd IGLC. Oslo, Norway. IGLC. Kjaerheim, G., 2005. Cleaner production and sustainability. J. Clean. Prod. 13 (4), 329e339. Koskela, L., 2000. An exploration towards a production theory and its application to construction. PhD Thesis. Helsinki University of Technology. Lapinski, A., Horman, M., Riley, D., 2006. Lean processes for sustainable project delivery. J. Constr. Eng. Manag. 132, 1083e1091. Lee, M.H., Lin, H.H., 2015. Evaluation of annual rainfall erosivity index based on daily, monthly, and annual precipitation data of rainfall station network in Southern Taiwan. In: International Journal of Distributed Sensor Networks, vol. 2015. Hindawi Publishing Corporation, pp. 1e15. Lewis, P., Hajji, A., 2012. Estimating the economic, energy, and environmental impact of earthwork activities. In: Construction Research Congress, Indiana, United States, May 21e23, ASCE. Magnusson, S., Lundberg, K., Svedberg, B., Knutsson, S., 2015. Sustainable management of excavated soil and rock in urban areas e a literature review. J. Clean. Prod. 93, 18e25. Maniquiz, M.C., Lee, S., Lee, E., Kong, D.S., Kim, L.H., 2009. Unit soil loss rate from various construction sites during a storm. Water Sci. Technol. 59, 2187e2196. Martinez, J.C., 1998. Earthmover-simulation tool for earthwork planning. In: Simulation Conference Proceedings, Washington, 13e16 Dec 1998. IEEE. Martínez, P., Gonz alez, V., Fonseca, E.D., 2009. Green-Lean conceptual integration in the project design, planning and construction. Revista Ingeniería de Conn J. 24 (1), 5e32. struccio Monden, Y., 1993. Toyota Production System: an Integrated Approach to Just-intime. Industrial Engineering and Management Press.

2208

S. Belayutham et al. / Journal of Cleaner Production 142 (2017) 2195e2208

Neto, R.S., Silvestre, B.S., Balloni, F., 2013. The impact of the cleaner production technologies in the mining productive chain. In: The Case of Padua-RJ. 4th International Workshop Advances in Cleaner Production, May 22-24, Sao Paulo, Brazil. NZQA (New Zealand Qualification Authority), 2015. Supervise Civil Construction Earthworks. Available At. http://Www.Nzqa.Govt.Nz/Framework/Explore/ Domain.Do?Frameworkid¼75329 (Accessed 18 February 2015). NZTA (New Zealand Transport Agency), 2010. Erosion and Sediment Control Field Guide for Contractors. NZ Transport Agency, Wellington. Ohno, T., 1988. Toyota Production System: beyond Large-scale Production. Taylor and Francis. Ooshaksaraie, L., Basri, N.E.A., Bakar, A.A., Maulud, K.N.A., 2009. An expert system prototype for minimizing soil erosion on construction site in Malaysia. Eur. J. Sci. Res. 33, 454e460. Pain, B., 2014. Glenvar Ridge Road - Erosion and Sediment Control Methodology. Woods, Auckland. Pampanelli, A.B., Found, P., Bernardes, A.M., 2014. A lean & green model for a production cell. J. Clean. Prod. 85, 19e30. Peurifoy, R.L., Oberlender, G.D., 2004. Earthwork and Excavation. Estimating Construction Costs, fifth ed. Mcgraw-Hill, New York. lez, V., Raftery, G., Orozco, F., 2014. Characterization of process Poshdar, M., Gonza variability in construction. J. Constr. Eng. Manag. 05014009(1)-05014009(10). Queiroz, G., Cobra, R.L.R.B., Guardia, M., Oliveira, J.A., Ometto, A.R., Esposto, K.F., 2015. The use of lean manufacturing practices in cleaner production: a systematic review. In: 5th International Workshop-Advance in Cleaner Production. Sao Paulo: Brazil. 20e22 May 2015. Rosenbaum, S., Toledo, M., Gonz alez, V., 2014. Improving environmental and production performance in construction projects using Value-Stream Mapping: case Study. J. Constr. Eng. Manag. 140 (11), 04013045(1)-04013045(11). Rother, M., Shook, J., 2009. Learning to See: Value-stream Mapping to Create Value and Eliminate Muda. Lean Enterprises Institute Incorporated, Cambridge, USA. Sayer, N.J., Williams, B., 2007. Lean for Dummies. Wiley Publishing Inc, Hoboken, New Jersey, USA. Shah, R., Ward, P.T., 2003. Lean manufacturing: context, practice bundles, and performance. J. Operations Manag. 21 (2), 129e149. Shaver, E., 2000. Low Impact Design Manual for the Auckland Region. Available at. http://Www.Aucklandcity.Govt.Nz/Council/Documents/Technicalpublications/ TP124%20Low%20Impact%20Design%20Manual%20for%20the%20Auckland% 20Region%20Chapter%201%20-%202000.Pdf (Accessed 6 January 2015).

Shehata, M.E., El-Gohary, K.M., 2011. Towards improving construction labor productivity and projects' performance. Alexandria Eng. J. 50 (4), 321e330. Stebbins, R.A., 2001. Exploratory Research in the Social Sciences. SAGE Publications, Inc., Thousand Oaks, pp. 2e18. Taylor, K., Field, M., 2007. Sustainable Construction: Reducing the Impact of Creating a Building. Available at. Http://Www.Branz.Co.Nz/Cms_Show_Download.Php? Id¼292a5866bacbf3aad00794c5f014c024f8f36a6d (Accessed 27 January 2015). Teh, S.H., 2011. Soil Erosion Modeling Using RUSLE and GIS on Cameron Highlands, Malaysia for Hydropower Development. (Masters). The School for Renewable Energy Science, Akureyri, Iceland. Thomas, H.R., Zavrski, I., 1999. Construction baseline productivity: theory and practice. J. Constr. Eng. Manag. 125 (5), 295e303. Trani, M.L., Bossi, S., Gangolells, M., Casals, M., 2016. Predicting fuel energy consumption during earthworks. J. Clean. Prod. 112 (Part 5), 3798e3809. UNEP (United Nations Environment Programme), 1996. Cleaner Production, first ed. United Nations Publication, Paris: France. Weese, D.R., 2007. Site inspections - can you stay out of jail?. In: 38th International Erosion Control Association Conference and Expo, Nevada: USA, 12-17 February 2007, pp. 488e493. Wischmeier, W.H., Smith, D.D., 1978. Predicting Rainfall Erosion Losses e a Guide to Conservation Planning. United States Department of Agriculture, Washington, D.C. Womack, J.P., Jones, D.T., Roos, D., 1990. The Machine that Changed the World. Rawson Associates, New York, USA. Womack, James, P., Jones, Daniel T., 1996. Lean Thinking. Simon and Schuster, New York. Wu, P., Low, S.P., 2012. Lean management and low carbon emissions in precast concrete factories in Singapore. J. Archit. Eng. 18, 176e186. Wu, Q., Wang, L., Gao, H., Chen, Y., 2012. Current situation analysis on soil erosion risk management in development and construction project. Adv. Mater. Res. 430e432, 1967e1969. Yao, H., Shen, L., Tan, Y., Hao, J., 2011. Simulating the impacts of policy scenarios on the sustainability performance of infrastructure projects. Autom. Constr. 20, 1060e1069. Yin, R.K., 2009. Case Study Research: Design and Methods. SAGE Publications. Zaini, A.A., Endut, I.R., 2014. Critical review on the theoretical framework and critical success factors in green construction. In: Hassan, et al. (Eds.), Proceedings of the International Civil and Infrastructure Engineering Conference. Springer-Verlag, Singapore, pp. 243e252.