Method for assessing the obsolescence of manufacturing equipment based on the triple bottom line

SC, Brasil *j.c.ferreira@ufsc.br Abstract Paper aims: This paper proposes a method for the assessment of equipment obsolescence, which includes the economic, environmental and social dimensions that compose the triple bottom line. Originality: This paper describes a novel method that allows a complete evaluation of the state of use of a machine, besides the traditional economic-functional analysis, as well as its application in a large electric machines manufacturing company. Research method: Analytic Hierarchy Process (AHP) multicriteria analysis is used, allowing a comparative and flexible assessment of the equipment. Main findings: Even if there is no economic justification for replacing the equipment considering the financial return on investment, the machine can be considered obsolete from a sustainable point of view, due to its inadequate environmental and social performance. Implications for theory and practice: Applying sustainability concepts in the manufacturing environment, more precisely in the machines on the shop floor, is very important to

In this context, this paper proposes a method to assess manufacturing equipment, based on the triple bottom line (TBL). The proposed method was applied for obsolescence assessment, using quantitative and qualitative sustainability indicators. This paper is structured as follows. In section 2 a review of the knowledge used in this research is presented, which includes the triple bottom line and performance indicators, product obsolescence, and the AHP multicriteria analysis method. Section 3 describes the proposed assessment method, including the hierarchical structure used in the study. In section 4 the results of the application of the method are presented and discussed in the production of stator and rotor blades of a product manufactured in a large company. Finally, section 5 presents the conclusions of the paper.

Triple Bottom Line (TBL)
The issue of sustainable development discussed by the Brundtland Commission (Brundtland, 1987) included the need to integrate aspects of environmental sustainability with economic and social aspects. Thus, it contributed to the definition of an important concept known as the Triple Bottom Line. This concept was initially established by Elkington (1999), which incorporates social, environmental and economic dimensions in the assessment of sustainable development. The author points out the need to include sustainability in companies' daily life, and also the importance of harmonizing the traditional economic dimension with environmental quality and social justice (Elkington, 1999). These three dimensions are also known as 3P's (People, Planet and Profit) (Kleindorfer et al., 2005).
The literature review performed by Morioka & Carvalho (2016) reports, based on the increase in the number of publications, that the inclusion of triple bottom line information in business performance indicators is quite significant in recent years. However, there are many opportunities for research, such as the search for more objective and tangible indicators for the social dimension.
Sustainability indicators provide a simple and affordable solution for rapid sustainability assessment. However, they have been rarely employed at the level of discrete manufacturing processes (Linke et al., 2013). Robinson (2013) reports that Caterpillar, in addition to other large corporations, has set sustainability targets to be met in the coming years to guide them towards sustainable development. However, understanding the limitations and opportunities in discrete manufacturing processes are critical aspects of the path to achieve these goals. According to Robinson (2013), overall sustainability indicators (e.g. energy consumption, water consumption and waste generation) are usually well quantified within the plant. However, information at the manufacturing process level is not readily available.

Equipment obsolescence
The function of a manufacturing equipment or system is related to its ability to perform the tasks for which the equipment was produced (Sénéchal, 2017). In this context, Panneerselvam (2012) explains the gradual loss of this ability or, in other words, the process of equipment deterioration in two main aspects. The first is related to the incapacity of the components of the equipment to perform their functions, while the second relates to technological obsolescence. Incapacity is caused by physical changes through use, fatigue, wear etc., resulting in increased maintenance costs. The second may have a direct impact on the product and/or operators involved, due to the use of technologically obsolete equipment. With regard to the operators, in addition to excessive manual work compared with more modern equipment, the equipment may not meet new safety recommendations, which have been modified over time. Concerning the product, technologically obsolete equipment can influence directly the quality of what is produced (Guo, 2017). Equipment obsolescence has been studied by many authors. Lawrence & Tanchoco (1990) carried out a study to assess economic obsolescence for a multi-equipment environment, based on investment, production, and marketing information. Hartman & Murphy (2006) explore if the economic life analysis for infinite-horizon equipment (including purchase, operating and maintenance costs), is a good policy for the finite-horizon problem, which occurs when companies only require an asset for a specified length of time, usually to fulfill a specific contract. Tan & Hartman (2010) expand equipment replacement analysis for the case where the horizon is not infinite or finite, but will last at least T s periods but may last as long as T l periods, due to uncertainty in the length of production runs or the temporary provision of services. Oliveira & Duque (2011) used partial differential equations to indicate the moment for equipment replacement. Pingle (2015) proposed a model to evaluate obsolescence in a multicriteria decision environment, based on quantitative and subjective factors.
A common concern among these works is the special attention given to the economic dimension and eventually the environmental dimension is included for the assessment of obsolescence. The study by Sénéchal (2017) deals with the application of concepts related to the triple bottom line in assessing the depreciation of a piece of equipment. The author proposed a procedure to verify the sustainability condition of a machine as a function of monitored indicators, to assist in choosing the correct maintenance policy. The decision for replacing the equipment is made by comparing obsolete equipment with possible replacement alternatives (Ravi, 2015). According to Ravi (2015), when the relevant factors used in the comparison are converted to a financial basis, the best alternative can be chosen by applying economic evaluation methods. Among the existing methods, in this work the payback period assessment was used, which represents the time of economic return on investment resulting from the gain from equipment exchange. However, by not taking into account the cost of money over time, the more traditional evaluation method known as simple payback was replaced by the discounted payback period (Jan, 2017). The assessment carried out in this study was based on the assessment presented by Cucchiella et al. (2016), represented in Equation 1.

( )
where DPBT is the Discounted Payback Time, t is the time of the cash flow, C t is the discounted cash flow, and r is the opportunity cost. DPBT corresponds to the moment when the sum of the discounted cash flows C t values becomes equal to zero. It should be pointed out that the initial period (t = 0) the first cash flow represents the total amount invested in the equipment. For the following periods, the cash flows are obtained through the cash savings corrected accordingly by the opportunity cost r, which represents the cost of money at each period of the cash flow.

Method of multicriteria analysis
There are several methods of multicriteria analysis that are applied in different areas of knowledge. Huang et al. (2011) carried out a study based on 312 papers published between 2000 and 2009 describing researches related to decision making in the field of sustainability. According to Huang et al. (2011), the significant use of the AHP (Saaty, 1987) (48%) can be explained by its ease of application and understanding by the decision maker, the greater availability of software, and the existence of user groups engaged in the subject.
More recently, Ho & Ma (2018) report that among the techniques that support decision making, AHP is the most often used and well known for the last 20 years, because of its great flexibility and wide applicability. The authors reviewed articles published between 2007 and 2016.
Other studies related to the assessment of sustainability parameters also highlight the advantages of the AHP method. Among them is the study carried out by Caputo et al. (2013), where the AHP method is used in a safety assessment of equipment, leading to a significant robustness of the ranking obtained by the method, besides the advantage of the consistency analysis of the judgment made by the decision maker. Ocampo (2015) provides a model for the assessment of a computational index for sustainable manufacturing, where the AHP method integrates adequately both objective and subjective evaluations made in the assessment of the sustainability of a company. Kluczek (2016) presents the application of the AHP method in the assessment of sustainability in manufacturing processes, and the method is appropriate due to the possibility of comparing established criteria, as well as their mathematical simplicity and flexibility. And Saad et al. (2019) report that the AHP method is easy to use, scalable, easily adjustable to fit many sized problems, and it is not data intensive.
According to Saaty (2008), the decision making process using the AHP multicriteria analysis method must be performed in an organized manner, according to the following steps: (a) Define the problem and determine the type of knowledge sought (overall objective of the assessment); (b) Structure the hierarchical levels of assessment. Starting from the top, with the ultimate objective of the assessment, following the alternatives from a broad perspective, through intermediate levels, to the lowest level where usually is the set with the alternatives; (c) Determine the two-way comparison matrices for each of the assessed alternatives, where each element at a level is used to compare with the elements at the level immediately below; (d) Use the priorities obtained from the comparisons to assign weights to the priorities immediately below in the hierarchical structure. At the end, for each element of the level below, add the weighted values and obtain the total values for that level. This process must continue until the last level of the hierarchical structure.
In order to perform the comparisons, it is necessary to use a scale of numbers that indicates how important or dominant one alternative is over another in relation to the criterion or property to which these alternatives are compared. This scale can be obtained through a quantitative or qualitative assessment (Saaty, 2008).

The proposed method for equipment assessment in the context of the triple bottom line
The proposed method for assessing the obsolescence of manufacturing equipment based on the triple bottom line was structured in seven steps, as shown in Figure 1. These steps are described below. 3.1.
Step 1 -Selection of the equipment that will be assessed Step in which the equipment to be assessed is chosen. It is important to note that more than two pieces of equipment can be assessed simultaneously, but the definition of obsolescence is made when comparing each one with the reference equipment.

Step 2 -Definition of the aspects in each dimension of the triple bottom line
The proposed method is based on the analytic hierarchy process (AHP) multicriteria analysis technique, and it is important to define the hierarchical levels. In this work the assessment has four hierarchical levels, as shown in Figure 2. The sustainability assessment (level 0) is carried out following an assessment of the three dimensions of sustainability (level 1). Each dimension is assessed through sustainability aspects (level 2) which, finally, are analyzed according to their respective sustainability indicators (level 3).
The choice of levels 2 and 3 was made based on the classification used in works such as Fan et al. (2010), Caetano et al. (2012) and Bhanot et al. (2015) in which, in order to facilitate the presentation and organization of information, the authors classified indicators within sustainability aspects. That is, in each of the three dimensions there may be some aspects, and, within each aspect, there are indicators as presented in this paper.
In order to create a list to guide the definition of sustainability indicators, a brief literature survey was carried out on this topic. Based on the review by Joung et al. (2013), which covers a large number of aspects and indicators, a search for some more publications in the last years related to the triple bottom line and applied to manufacturing processes was performed. In summary, the main aspects within each of the three dimensions of sustainability are presented in Table 1. Economic Emission, waste and effluents Worker training and satisfaction Table 2. Indicators for the economic dimension.

Economic Aspects Sustainability Indicators
Cost Cost of direct and indirect labor, cost of maintenance, cost of energy consumption, materials and supplies, cost of inventory, etc.

Productivity
Cycle time, production rate, overall equipment effectiveness (OEE), availability of equipment, quantity of material produced, etc.

Quality
Relation between non-conforming parts and good parts, capacity index and process performance (P p and P pk ), etc.

Guarantees and duties
Cost of treatment of fluids, cost of treatment of waste, cost resulting from worker indemnities, etc.  Table 3. Indicators for the environmental dimension.

Environmental Aspects Sustainability Indicators
Consumption of inputs Consumption of electric energy, gases, water, oils, etc.
Material usage Raw material used, consumption of tools, filters, use of restricted materials, etc.
Occupied area Total area required for installation of the machine and accessories, area required for the material being processed, etc.
Emissions, waste and effluents Quantity of oil discarded, level of water loss, emission of CO 2 and other gases, emission of toxic substances, etc. In order to be used as a guideline for the choice of alternatives to be compared, a list with the main indicators used for each of the sustainability aspects was explored along with the references in Table 1. Table 2 presents some indicators for the economic dimension, separated between aspects, while Table 3 refers to the environmental dimension. Finally, Table 4 shows the list of indicators for the social dimension.
Some aspects or indicators presented in the evaluated studies are not listed because they are not focused on the equipment but in the company. For example, aspects related to community development, distance from suppliers to the plant, noise level outside the company, among others. Based on the information in Tables 1-4, together with the definition of the hierarchical levels of Figure 2, the general list of aspects and indicators for assessment of the obsolescence of manufacturing equipment was set up as shown in Figure 3.
The method also allows the entry of new aspects and indicators, according to one of the three dimensions of sustainability. Table 4. Indicators for the social dimension.

Social Aspects Sustainability Indicators
Labor intensity Energy consumed during machine operation, energy consumption in setup, etc.
Occupational health Level of noise and vibration, degree of unhealthiness; problems of ergonomics, lighting and ventilation, etc.
Training and satisfaction Degree of instruction required for the operator, hours of training, level of satisfaction to perform the task, career development, etc.
Job safety Number of accidents at work, risk of electric, mechanical and chemical accidents, danger degree of the task, etc.  • Similar indicators can be included in aspects listed in different dimensions. Thus, the impact of this indicator is being assessed from different perspectives. For example, electricity consumption is an important cost indicator for the economic dimension and also a source of significant environmental impact depending on the local energy matrix; • Quantitative or qualitative indicators can be assessed.

Step 4 -Collection of data related to chosen indicators
Based on the chosen indicators, information is collected. Some observations are needed at this stage: • Quantitative indicators should always be compared on a common basis and representative of actual production. For this, preferably, the main unit should be normalized by the quantity of parts produced. For example, kWh/part or mg/part; • The qualitative indicators will be assessed based on the fundamental scale of Saaty (2008) presented in Table 5.
It should be mentioned that the classic AHP method should be followed by comparing the criteria to ultimately obtain a global function that promotes the definitive classification of all levels and alternatives evaluated. However, the method proposed in this study uses another function to promote a definitive classification, as shown in the next steps.

3.5.
Step 5 -Determining the weight of dimensions and aspects of sustainability The method requires the determination of weights for dimensions (level 1) and aspects (level 2). The weight of dimensions (dw) is obtained automatically according to Equation 2, while the aspect weight (aw) is obtained according to Equation 3. The weight of dimensions is based on the number of aspects related to the dimension, compared with the total number of aspects. Likewise, the weight of the aspects is obtained by dividing the number of indicators of the aspect by the total number of dimension indicators of which the aspect is part. Thus, a possible aspect of which many indicators are assessed should have a larger weight since several items are being analyzed. Regardless of this weight suggestion, it is up to the assessor to evaluate the weights, just by noting that the sum of them has to be equal to 1.00 (or 100%).

Number of Aspects in Dimension Dimension Weight dw Total Number of Aspects =
(2) ( )

Number of Indicators in Aspects AspectWeight aw Number of Indicators in Dimension =
(3) 3.6.
Step 6 -Analysis of the machines At this stage the equipment is assessed and the value of the Sustainability Index for Machine Obsolescence (SIMO) is determined. For this, the comparison by the AHP method for each of the chosen qualitative indicators is used as the basis, and the quantitative indicators are normalized. This normalization is done for each machine i, according to Equation 4. An important issue concerns the nature of the indicators, since, for each indicator, a larger value indicates that the equipment has a higher degree of obsolescence. Thus, in the case where a lower value of an indicator represents a machine with greater obsolescence (for example, "production rate"), the reciprocal of the value for each equipment must be obtained before normalization, according to Equation 5  Extreme importance The evidence favoring one task over another is of the highest possible order of affirmation

Reciprocals of above
If task i has one of the above non-zero numbers assigned to it when compared with task j, then j has the reciprocal value when compared with i A reasonable assumption 1.1 -1.9 If the activities are very close It may be difficult to assign the best value, but when compared with other contrasting activities the size of the small numbers would not be too noticeable, yet they can still indicate the relative importance of the activities. 3.7.
Step 7 -Assessment of results The last step of the method corresponds to an assessment of the results obtained. The machine that has the highest SIMO value, calculated in step 6, has greater obsolescence based on the triple bottom line. As a guideline, a percentage analysis of the value of each of the machines can be made to facilitate the understanding of the result, as well as a comparison in each dimension to expose the areas with the greatest problems.
The fact of having greater obsolescence in terms of sustainability does not mean that the machine should be replaced. This diagnosis is made by comparing a Sustainability Index for Machine Obsolescence (SIMO) with the ratio between the discounted payback (for example, for the purchase of new equipment or even a retrofitting of the equipment under analysis) and the company's payback policy. Figure 4 illustrates how this comparison is made. In this way, it is possible to carry out another type of analysis for situations where there is no economic justification for the approval of the investment, i.e. for situations where the calculated discounted payback period (X months) is greater than the return on investment policy (Y months). In this case, the assessor can analyze which combinations of weights of the dimensions (dw) represent the obsolescence of the machine "i" compared with the reference machine "ref" as a function of the ratio between ( env soc dv dv + ) and SIMO, compared with the ratio between payback and payback policy (X/Y). eco dv is not considered in the numerator because the assessment of the return on investment period is an indirect representation of the economic dimension that should not be reconsidered.

Results
The proposed method was applied in a large company that manufactures electric machines (motors, generators and transformers) located in the state of Santa Catarina, southern Brazil. For the manufacture of electric motors' laminations, as well as small electronics components, large-scale production equipment is used, known as high-performance lines with a high production rate (stamping speed), also called high speed stamping press lines (Hoffman & Kasparbauer, 1998). The company has a production line with more than 15 high speed presses that process together more than 300 tons of steel per day. According to step 1 of the method, two machines were chosen from the same manufacturer, whose main data are shown in Table 6.  Steps 2 and 3 were performed with the help of the people directly involved with this manufacturing process in the company. No additional aspects were added to the already pre-defined 12 aspects (shown in Table 1), but some indicators were included or simply adapted. The units of the indicators should follow magnitudes already regularly used. It was also decided that some indicators would be assessed in a qualitative way, and they receive the word "qualitative" in the field related to their unit as shown in Table 7.
Some indicators related to progressive stamping have not been listed because they relate only to the tool (progressive stamping die) and not to the presses themselves. For example, the indicators referring to the raw material and its use depend on the design solutions for the tools. No indicator of water consumption was included, since it is included in the consumption of cutting fluid. Table 7 shows all the indicators chosen for the evaluation of the two presses, and some details of the dimensions are given below.
(a) Economic Dimension: • The "maintenance costs" indicator includes the costs related to maintenance work, exchanged parts and stopped machine due to corrective maintenance.
• The "production costs" indicator represents, in addition to the cost of direct and indirect labor, the costs of electricity, used tools, machine depreciation, etc.
• For the indicators "production rate" and "setup time" the values referring to the same parts produced in the presses being assessed should be compared, in order to avoid the influence of the variation of the tool.
• The "percentage of nonconformance" is related to the number of parts outside the specification that will have to be rejected, as well as to those that need rework.
(b) Environmental Dimension: • The "tooling" indicator represents the tool wear found on each of the different machines being assessed. It is measured at each regrinding of the stamping tool.
• For the assessment of indicators related to "material in process", in the aspect "occupied area", it should be based on the same type of product so that very different sizes of parts are not assessed.
(c) Social Dimension: • The "level of instruction" indicator is related to the educational content required for the operation of the high speed presses.
• For the indicator "degree of motivation for the task" it is intended to analyze the interest and motivation of the operators due to the operational difference and complexity of the presses and their accessories.
• The "number of accidents" indicator should cover all situations that cause (or not) removal from the task.
Following the proposed method, Step 4 corresponds to the collection of information. The quantitative indicators were measured and/or assessed based on the data collected by the machine monitoring system, and the qualitative indicators were assessed based on the knowledge of the company's employees. The last two columns of Table 7 present the result of these measurements and/or assessments.
The description of the indicators was presented in step 3, but the following comments are necessary for a better understanding of the assessment: • The assessment period was 18 months; • For the "production rate" and "setup time" indicators, the values referring to the same progressive stamping dies used on these two machines were compared; • The "percentage of nonconformance " indicator was obtained based on an indicator already monitored by the company; • The "quantity of CO 2 emitted" was estimated based on the electric energy consumption of each piece of equipment; • With regard to the assessment of a quantitative indicator whose highest value corresponds to a greater obsolescence of the press, an example of calculation using Equation 4 for the indicator "maintenance" of the aspect "costs" is shown in Table 8.
• In the case of a quantitative indicator whose lower value corresponds to a greater obsolescence of the press, an example of calculations using Equations 5 and 6 for the indicator "rate of production" (of aspect "productivity") results in the values shown in the last two columns of Table 9.
• In the case of a qualitative indicator in which its larger value corresponds to a greater obsolescence of the press, the example of the indicator "risk of accidents in the operation" is shown. In this case, Press A has a greater risk of causing accidents in the operation than Press B. Thus, the priority vectors of the AHP method are determined, which are considered the normalized values (Equation 4), and they are shown in the last column of Table 10.
• In the case of a qualitative indicator whose lower value corresponds to a greater obsolescence of the press, the example of the indicator "ergonomics in setup" is presented. In this case, it was observed that Press A is somewhat less ergonomic than Press B. Table 11 shows the values assigned to the presses as well as priority vectors applied. Equations 5 and 6 are applied to determine the normalized values.
The next step corresponds to the determination of the weights of aspects and sustainability indicators (Step 5). Using Equations 2 and 3 the values shown in Table 12 are obtained.    ) are also presented.
Step 7 begins with the sustainability assessment in each dimension, where it was observed that press A presented the worst result for the economic dimension (56% worse), followed by the social dimension (40%) and 21% for the environmental dimension in comparison with the reference press. In the environmental dimension, the difference was less because press A (20 years older than press B) required a smaller installed area and had less tool wear, since it ran at speeds much lower than the speeds of the newer press.
In the overall assessment press A's Sustainability Index for Machine Obsolescence is 67% higher than the value found for press B. This fact was already expected due to the difference of more than 20 years between these machines and the different installed resources.

Following
Step 7, the obsolescence of press A was analyzed based on the triple bottom line, with the help of some company information (Table 14).
Applying Equation 1, the discounted payback time is equal to 76 months, which is 27% higher than the payback policy adopted by the company for this type of investment. However, as a sequence of step 7 different weight combinations were assigned to the dimensions using Equation 10 in the percentage form and considering press B as reference. As shown in Table 15, it can be observed that a weight of 40% for the economic dimension, 30% for the environmental dimension and 30% for the social dimension results in the press being diagnosed as obsolete from the sustainability standpoint, since the value was above 27%.  There are other combinations of weights which satisfy the comparison of Equation 10. In order to perform a sensitivity analysis of the influence of the variation of the dimensions weights, Figure 5 shows in the gray highlighted region options of combinations of weights for environmental and social dimensions that result in press A being considered obsolete. The weight of the economic dimension is not explicit, but it was used as the complementary value of the other dimensions to reach the sum of 100%. In addition, blank cells represent not possible combinations, where the sum of the weights exceeds 100%.
Finally, the final assessment shows that even if there is no economic justification for replacing the press A, based on the assessment of the financial return on investment, the press can be considered obsolete from a sustainable point of view due to its environmental and social performance.

Conclusions
The proposed method sought to provide the assessment of equipment based on indicators related to the triple bottom line. The method allows the assessor to include new criteria for the assessment, as well as to analyze qualitative information about the equipment under analysis.
As for the assessment of obsolescence, the study presented a broader view of the traditional analysis of the economic performance of machines, providing an expansion for indicators related to the environmental and social dimensions. In this way, the assessor is allowed to decide on the replacement of equipment based on an assessment in which sustainability is taken into account.
With regard to the assessment of high speed presses, an unexplored study of such equipment was presented, which are machines that work in 24-hour production daily in a large company. Several indicators that are considered in this work have an effective value only when assessed in a large-scale production situation.
There is no restriction for the application of the proposed method in other types of machines and, therefore, as a future work, it is suggested the application of the method in different types of equipment. It is also recommended the development of software that allows the creation of a database with the collected information, as well as a method to eliminate human intervention in the analysis of the weights that indicate the obsolescence of the machine in the sustainability point of view. In addition, future work may contemplate different techniques of financial analysis of the investment, as well as include variable cash savings over time due to variations in maintenance cost, variations in productivity, and costs resulting from nonconformances in manufactured parts.