Knowledge-Inference-Based Intelligent Decision Making for Nonferrous Metal Mineral-Processing Flowsheet Design
Round 1
Reviewer 1 Report
Comments and Suggestions for AuthorsThis research proposes an intelligent decision support framework based on knowledge graphs and large model reasoning aimed at optimising the processing of non-ferrous metal ores. By combining natural language processing and graphical techniques, the framework is able to provide customised process optimisation recommendations based on the genetic characteristics of the ore and the processing needs. However, despite the large contribution to the theoretical innovation of ore processing, there are still some shortcomings. Revisions are required for submission.
- In the genetic characterisation of ores presented in the paper, how to ensure that the causal relationship between ore properties (e.g. mineral composition, structure and their physicochemical changes during treatment) and the final process design is effectively quantified?
- How does a knowledge graph-based ore treatment inference system handle the complex reactions and interactions between minerals in the ore? Does it take into account the impact of chemical reactions on the final treatment results?
- Is the system able to make real-time adjustments to optimise the treatment process based on dynamic changes in the ore (e.g., changes in ore grade during the mining process)?
- How does the Dynamic Decision Support Framework (DDSF) mentioned in the paper balance the weights of the objectives when dealing with uncertainty and multi-objective optimisation problems?
- Is the framework capable of self-learning and continuously optimising process decisions based on new ore processing data?
- The authors should reorganise their conclusions.
Comments for author File: Comments.docx
Author Response
Please see the attachment.
Author Response File: Author Response.docx
Reviewer 2 Report
Comments and Suggestions for AuthorsDear Authors,
The research lacks model validation, i.e. reliable evaluation indicators or a reliable way of showing that the model actually achieves the intended goals. How accurate is the model? The authors only confirm that they tested the models in three mines, but they do not show measurable numerical effects, for example, by what percentage of the metal recovery was increased. There are only recommendations on how to conduct the process, e.g., crushing or flotation, but this knowledge is obvious and known to every qualified mineral processing engineer. The model suggests obvious solutions. Research by the authors does not reliably confirm that the model recommendations will be beneficial and bring a positive technological effect.
If the authors expand their research with the conclusions they have formulated in this publication, then the future article will be of great scientific importance. The accuracy of the model and its reliable validation must be given. The authors should start the validation from laboratory conditions, because it is very difficult to carry out the model validation under industrial conditions, because each introduction of major technological changes is expensive and carries risk. In its current form, it is more of a popularizing article that provides concepts for the operation of an intelligent decision support model in mineral processing, but it is not validated. The authors propose chatGPT for the mineral processing industry. It is an extremely valuable model, but at this stage of development, its recommendations/guidance are obvious and unverified. In its current form, the article has no scientific value, further research is needed to verify the effectiveness of the model.
I express great appreciation to the authors for working on such an important, intelligent decision support tool, which is extremely needed in mineral processing, because these processes are variable, dynamic, subject to disruption, synergistic, and difficult to control.
Author Response
Please see the attachment.
Author Response File: Author Response.docx
Round 2
Reviewer 1 Report
Comments and Suggestions for AuthorsAccept in present form
Reviewer 2 Report
Comments and Suggestions for AuthorsI am satisfied with the explanations of the authors and the changes introduced in the manuscript. The authors have validated their intelligent model with reliable laboratory tests and the disclosed supplement. Provides fundamental principles of mineral processing depending on the characteristics of the ore. The authors have extensive mineral processing knowledge and are aware of the fundamental principles. Industrial validation remains to be solved, which is not an easy task. I would like the authors to improve their intelligent model because it is very necessary for mineral engineering. The manuscript, after the improvements, is fully suitable for publication.
Congratulations and best wishes.