Developing Expert Systems for Improving Energy Efficiency in Manufacturing: A Case Study on Parts Cleaning
Round 1
Reviewer 1 Report
Comments and Suggestions for AuthorsThe esteemed authors have conducted a significant and practical study on the development of expert systems for improving energy efficiency in manufacturing. Through data-driven regression models, they have successfully identified inefficient parameter settings and calculated achievable energy savings. However, before publication, it is recommended that the following issues be addressed to ensure the accuracy and integrity of the research:
1. The authors should consider expanding the literature review to include not only directly related studies but also indirectly related research such as the application of expert systems in other domains, methods for energy efficiency assessment, and the utilization of fuzzy logic in decision support systems. This broader perspective will strengthen the theoretical foundation of the work.
2. The authors should provide a clear description of the data sources, collection methods, and preprocessing steps. This will ensure the reproducibility and verifiability of the research. Details such as data sampling techniques, cleaning procedures, and any transformations applied should be included.
3. The authors should elaborate on the construction process of the data-driven regression models, including feature selection, model training, and parameter tuning. Furthermore, the design and construction of the fuzzy rule base as well as the model validation and evaluation methods should be described in detail. This will provide a robust foundation for the model's performance.
4. A thorough description of the parts cleaning process, including the current equipment used, process parameters, and energy consumption, should be provided. This will help readers understand the context and relevance of the study.
5. The authors should include a screenshot or mockup of the interface of the developed expert system to visually demonstrate the research outcomes. This will enhance the comprehensibility of the work.
6. A detailed explanation of how the expert system is applied to the case study, including the identification of inefficient parameter settings, calculation of achievable energy savings, and prioritization of actions based on the fuzzy rule base, should be provided.
7. A thorough analysis and discussion of the expert system's output, comparing energy consumption before and after implementation, should be conducted. This will help assess the effectiveness of the system.
8. The authors should clearly summarize the main findings of the paper, highlighting the effectiveness of the expert system in improving energy efficiency, as well as any limitations and challenges encountered.
9. The authors should emphasize the academic contributions and novel aspects of the work, such as new research methods, models, or application cases. This will demonstrate the significance and impact of the research.
10. The authors should discuss potential future research directions, such as applying the expert system to other production areas or integrating artificial intelligence techniques for intelligent decision support. This will provide a roadmap for future research.
Comments on the Quality of English LanguageMinor editing of English language required
Author Response
Comment1: The authors should consider expanding the literature review to include not only directly related studies but also indirectly related research such as the application of expert systems in other domains, methods for energy efficiency assessment, and the utilization of fuzzy logic in decision support systems. This broader perspective will strengthen the theoretical foundation of the work.
Response1: We strongly agree with this comment. For this reason, we wrote a review paper at the end of last year. That paper systematically reviews state-of-the-art approaches of ESs aimed at improving energy efficiency in industry, with a focus on manufacturing. The literature search yields 1692 results, of which 54 articles published between 1987 and 2023 are analyzed in depth. These publications are classified according to the system boundary, manufacturing type, application perspective, application purpose, ES type, and industry. Furthermore, we examine the structure, implementation, utilization, and development of ESs in this context. Through this analysis, the review reveals research gaps, pointing toward promising topics for future research. Unfortunately, the paper has been and still is in the decision-making process at another journal for several months.
Comment2: The authors should provide a clear description of the data sources, collection methods, and preprocessing steps. This will ensure the reproducibility and verifiability of the research. Details such as data sampling techniques, cleaning procedures, and any transformations applied should be included.
Response2: We have expanded the description: "Since outliers have no relevant influence on the average active power (which is to be represented by the regression models), no data cleaning or transformation is required. - page 8, section 4.2, line 268-269. The electrical power is measured with Janitza UMG 604 and Janitza 20CM metering devices at a sampling rate of 1 Hz and transmitted via Modbus TCP." - page 8, section 4.2, line 269-271.
Comment3: The authors should elaborate on the construction process of the data-driven regression models, including feature selection, model training, and parameter tuning. Furthermore, the design and construction of the fuzzy rule base as well as the model validation and evaluation methods should be described in detail. This will provide a robust foundation for the model's performance.
Response3: We apply basic regression models (linear, quadratic and cubic functions) to keep the complexity as low as possible. The models are applied straight to the average active power without the need for features or parameter tuning. The script for model creation and evaluation using R^2 and RMSE has been published in the referenced Git repository (reference number 37). We have now also uploaded the data (average active power of all consumers) with which the models were created to Github. The fuzzy rule base was constructed by the knowledge engineer and energy manager as described, based on the reference [15]. The rule base proved to be sufficient in the evaluation, although iterative adjustments and refinements would be possible according to the methodology presented.
Comment4: A thorough description of the parts cleaning process, including the current equipment used, process parameters, and energy consumption, should be provided. This will help readers understand the context and relevance of the study.
Response4: The specific machine type with data sheet can be found in the reference [38]. The process parameters considered are listed in Table 2. For better comparability, we consider the electrical power instead of the energy. For the Pareto prioritization (Figure 4), however, we have also included the observed time period (2 hours) - page 7, section 4.1, line 248.
Comment5: The authors should include a screenshot or mockup of the interface of the developed expert system to visually demonstrate the research outcomes. This will enhance the comprehensibility of the work.
Response5: The most relevant parts of the interface are included as figures in the paper (e.g. Figure 7 and Figure 8). The complete expert system, which is built in Jupyter Notebook, can be accessed via the referenced Git repository (reference number 37)
Comment6: A detailed explanation of how the expert system is applied to the case study, including the identification of inefficient parameter settings, calculation of achievable energy savings, and prioritization of actions based on the fuzzy rule base, should be provided.
Response6: Thank you for pointing this out. We added the following part: "When applying the ES in the case study, the current parameter settings are read from the PLC via OPC UA during the operation of the TPCM. The calculation of the savings potential $\Delta \overline{p}_{i}$ and the optimization potential $o_{CPP}$ is carried out according to equations (2) and (3). With these two values, the ES performs the prioritization using the fuzzy output $Z_{CPP}$." - page 12, section 4.6, line 381-384.
Comment7: A thorough analysis and discussion of the expert system's output, comparing energy consumption before and after implementation, should be conducted. This will help assess the effectiveness of the system.
Response7: Thank you for this important point! We added the following part: "According to Table 4, 4.81 kW can be saved solely by implementing the measure with the highest priority - lowering T_fluid from 50 °C to 40 °C. This would result in an absolute saving of 9.62 kWh and a relative saving of 19.95 % for the two-hour reference period as indicated in Figure 4. The sum of all achievable savings amounts to 10.93 kW. For the reference period, this would correspond to an absolute saving of 21.87 kWh and a relative saving of 45.35 %." - page 14, section 4.6, line 405-410.
Comment8: The authors should clearly summarize the main findings of the paper, highlighting the effectiveness of the expert system in improving energy efficiency, as well as any limitations and challenges encountered.
Response8: To emphasize the benefits of the ES, we have added the following part: "In the case study shown, the ES reveals a considerable energy saving potential of up to 45.35 % compared to the reference scenario." - page 14, section 5, line 416-417.
One main limitation was ponited out: "But it should be emphasized that the present system is not able to determine the effects on the production result." - page 15, section 5, line 428-430.
The main challenge is also pointed out in section 5: "During the implementation of the methodology for the use case shown, the authors also found that the effort required for analytical modelling is significantly higher than for data-driven electrical modelling." - page 15, section 5, line 433-435.
Comment9: The authors should emphasize the academic contributions and novel aspects of the work, such as new research methods, models, or application cases. This will demonstrate the significance and impact of the research.
Response9: We have added a summary of the developed methodology: "This article presents a systematic approach to the development of ESs for production machines and demonstrates it using a TPCM. Three phases are addressed for this process. The conceptual design phase and the implementation phase comprise the planning and execution necessary for the development of ESs. The partial outcomes and the final product - the ES - are applied and validated." - page 15, section 5, line 412-416. Further the novelty of this work is mentioned on page 2-3, section 1 , line 88-95.
Comment10: The authors should discuss potential future research directions, such as applying the expert system to other production areas or integrating artificial intelligence techniques for intelligent decision support. This will provide a roadmap for future research.
Response10: A few future research topics are adressed in section 5: "Adjustments to the simulation model are required to model potential direct consequences on the conveyed metal parts. In particular, the static thermal masses must be complemented with dynamic masses, which are put through by the conveyor belt. - page 15, section 5, line 430-432.
"...the methodology presented will be generalised so that it is not limited to individual production machines, but can be applied to several machines at a higher level of abstraction. It is also possible to consider energy flexibility in addition to energy efficiency in order to utilise the ES for demand response applications. - page 15, section 5, line 437-439.
Reviewer 2 Report
Comments and Suggestions for AuthorsThis paper addressed the challenge of energy efficiency in industrial manufacturing processes. The authors developed an expert system, with a data-driven approach using regression models and a fuzzy rule base, aimed at identifying and optimizing inefficient parameter settings in parts cleaning processes to achieve energy savings. The paper presents an interesting concept for developing expert systems to improve energy efficiency in manufacturing, but significant revisions are needed to address the bellow issues.
(1) Provide more details on the methodology used to develop the expert system. Explain the data collection process, the selection of regression models, and the fuzzy rule base in a more thorough manner. This will increase the reproducibility and rigor of your work.
(2) While the case study provides some insight into the potential energy savings, a more comprehensive evaluation of the expert system's performance, including comparisons to other approaches and statistical significance tests, would be beneficial.
Author Response
Comment1: Provide more details on the methodology used to develop the expert system. Explain the data collection process, the selection of regression models, and the fuzzy rule base in a more thorough manner. This will increase the reproducibility and rigor of your work.
Response1: We have expanded the description for the data collection process: "Since outliers have no relevant influence on the average active power (which is to be represented by the regression models), no data cleaning or transformation is required. - page 8, section 4.2, line 268-269. The electrical power is measured with Janitza UMG 604 and Janitza 20CM metering devices at a sampling rate of 1 Hz and transmitted via Modbus TCP." - page 8, section 4.2, line 269-271.
We apply basic regression models (linear, quadratic and cubic functions) to keep the complexity as low as possible. The models are applied straight to the average active power without the need for features or parameter tuning. The script for model creation and evaluation using R^2 and RMSE has been published in the referenced Github repository (reference number 37). We have now also uploaded the data (average active power of all consumers) with which the models were created to Github. The fuzzy rule base was constructed by the knowledge engineer and energy manager as described, based on the reference [15]. The rule base proved to be sufficient in the evaluation, although iterative adjustments and refinements would be possible according to the methodology presented.
Comment2: While the case study provides some insight into the potential energy savings, a more comprehensive evaluation of the expert system's performance, including comparisons to other approaches and statistical significance tests, would be beneficial.
Response2: We have added the following part to give a more concrete description of the results for the case study: "According to Table 4, 4.81 kW can be saved solely by implementing the measure with the highest priority - lowering T_fluid from 50 °C to 40 °C. This would result in an absolute saving of 9.62 kWh and a relative saving of 19.95 % for the two-hour reference period as indicated in Figure 4. The sum of all achievable savings amounts to 10.93 kW. For the reference period, this would correspond to an absolute saving of 21.87 kWh and a relative saving of 45.35 %." - page 14, section 4.6, line 405-410.
We focus on the ES's approach because it enables to overcome the mentioned barriers for improving energy efficiency as well as contributing to knowledge transfer. Other intelligent approaches are usually not capable of achieving this due to their black box nature and lack of explanation regarding the solution finding process.
Reviewer 3 Report
Comments and Suggestions for AuthorsThe manuscript presents a systematic approach to the development of ESs for production machines,and it has been developed in the application of parts cleaning. The work has industrial relevance, and the method has potential to improving energy efficiency.
The following factors need to be addressed for publication:
1) In Section 3.2, the ultimate goal and the exchange of data between concept and implementation are missing from the methodological framework.
2) In Section 4, the presentation of the contents lacks organization, and the proposed methodology corresponds to the expert system and indicates the model developed.
3)Are there tests to validate the proposed methodology?
4) In Section 4.5, it is recommended to use image or flowcharts to describe the integration.
5) In Seciton 5, more data are needed to elaborate on the conclusions.
Author Response
Comment1: In Section 3.2, the ultimate goal and the exchange of data between concept and implementation are missing from the methodological framework.
Response1: Thank you for pointing this out. We have tried to make the connection clearer: "The conceptual design phase involves planning the steps that are necessary to develop the ES. These steps are realized using different methods within the implementation phase. Data, information and knowledge are acquired during the implementation. As a result, artifacts are created that need to be validated and, if necessary, adjusted." - page 5, section 3.2, line 164-168.
Comment2: In Section 4, the presentation of the contents lacks organization, and the proposed methodology corresponds to the expert system and indicates the model developed.
Response2: We added the following part: "The case study considers the cleaning of metallic guide discs for gearboxes at the ETA research factory, which are contaminated with cutting oil." - page 7, section 4, line 234-235
Comment3: Are there tests to validate the proposed methodology?
Response3: The only way to validate the proposed methodology is through field studies. In this paper we demonstrate its applicability to parts cleaning. In further publications we plan to demonstrate the methodology on other production machines.
Comment4: In Section 4.5, it is recommended to use image or flowcharts to describe the integration.
Respsonse4: We strongly agree with this statement. We are currently working on a follow-up publication that focuses on the software implementation of the methodology presented. The follow-up publication will provide an ES shell for rapid and simplified implementation of the methodology described. This architecture will then also be visualized.
Comment5: In Seciton 5, more data are needed to elaborate on the conclusions.
Respsonse5: We have made a few additions in this respect. To clarify the benefits of the expert system: "According to Table 4, 4.81 kW can be saved solely by implementing the measure with the highest priority - lowering T_fluid from 50 °C to 40 °C. This would result in an absolute saving of 9.62 kWh and a relative saving of 19.95 % for the two-hour reference period as indicated in Figure 4. The sum of all achievable savings amounts to 10.93 kW. For the reference period, this would correspond to an absolute saving of 21.87 kWh and a relative saving of 45.35 %." - page 14, section 4.6, line 405-410.
To point out limitations and challenges encountered: "But it should be emphasized that the present system is not able to determine the effects on the production result." - page 15, section 5, line 428-430.
To emphasize the academic contributions and novel aspects of the work: "This article presents a systematic approach to the development of ESs for production machines and demonstrates it using a TPCM. Three phases are addressed for this process. The conceptual design phase and the implementation phase comprise the planning and execution necessary for the development of ESs. The partial outcomes and the final product - the ES - are applied and validated." - page 15, section 5, line 412-416.
Round 2
Reviewer 1 Report
Comments and Suggestions for AuthorsThe author has fully responded to the issues I raised, and I currently have no further questions. I recommend acceptance.
Comments on the Quality of English LanguageMinor editing of English language required
Reviewer 2 Report
Comments and Suggestions for AuthorsThe authors have responded to the questions raised by the reviewer and resolved them. The current version of the manuscript is acceptable.
