SaaMES: SaaS-Based MSA/MTA Model for Real-Time Control of IoT Edge Devices in Digital Manufacturing
Round 1
Reviewer 1 Report
The article is presenting a SaaS-based MES approach based mainly in the Microservices Architecture. The paper is well written. The authors provide a lot of background information. Nevertheless, both in introduction and in related work sections they are focusing on the SaaS and MSA approaches rather than to the approaches in the MES that allowed it to be digitized and fully automated. There are a lot of work in vertical integration of industrial systems where approaches such MSA are applies in the MES.
Further, they have presented a complex architecture with various components and aspects. The authors give a few details of the components of the architecture, and they are focusing mainly on the MTA approaches. For example, what is the role of the Digital Twin Agent? Also, the authors have decided to provide each microservice per customer, this decision needs more elaboration. In bibliography, there are other methods that are applied in separating the functions on the MES, such as the production processes and the products.
The section 3.4 is dedicated to the real-time aspect with IoT devices. In a manufacturing environment there are not only IoT devices and of course there are a variety of networks, industrials, and common networks. The authors are presenting an approach, very interesting indeed, but it seems very dedicated, they must elaborate how can these apply in a real manufacturing environment.
Finally, the authors must consider the real challenges in a manufacturing environment, apart from the time constrains and the resilience, there are more complex task that must be done in the shop floor in cooperation with MES, for example the reconfiguration of the machines to support different production lines and the machine maintenance. How will these affect their MSA architecture?
Author Response
Please see the attached for your review and comments.
Author Response File: 
Author Response.pdf
Reviewer 2 Report
The paper proposed the optimal method for MEA using MTA and MSA. the main purpose of the research is to reduce the complexity and cost. More testing is required to prove the efficiency of new architecture but the transitioning process to MEA and MSA is clearly presented.
Author Response
Please see the attached for your review and comments.
Author Response File: Author Response.pdf
Reviewer 3 Report
The submission makes the bold promise to present the concept and the experimental evaluation of a “an architecture of SaaS-based MES (Manufacturing Execution System) with MSA (microservices architecture) and MTA (Multi Tenancy Architecture)” for use in plants where some application services are latency sensitive and therefore need to be hosted nearer to the Edge than the centre of the Cloud. In fact, what the submission actually delivers gets nowhere near the target, which was already difficult to make sense of in the first place.
The biggest drawback and disappointment of this submission is that its contents for the most part skim over, in a shallow and superficial tone, all of the topics, arguments, claims, and discussions that it touches upon.
The introduction (section 1) and the related work (section 2) typify that defect by gluing together passages drawn from bibliography without however providing them with decently sound contextual justification, definition, and justification. Section 3, which is supposed to be the central part of the submission, is for the most part a collection of unsubstantiated claims accompanying block diagrams (which are what the authors consider “architectures”) that scarcely define the notions that they use, for example what is “tenancy” and how is “multi-tenancy” attained, what microservices are and how and why they serve the authors’ intent. This section then plunges into the element of real-time control, which, while expectedly central to the work, makes it first appearance at page 14 of 20 (section 3.4), with 20 lines of text that prelude to experiments (section 4) completely vacant of rationale and traceability to initial goals and requirements. The results that the authors present do not show novelty or special significance, in the general sense, and can scarcely be transferred as lessons learned to other contexts owing to their disappointingly shallow nature (see for example Figure 19, which shows “an increase by 33.1%” (hardly traceable to individual cost components) in a service delivered in a “traditional Cloud-centric manner” in comparison to an Edge-centred solution.
On the whole, the submission is naïve in discourse, exceedingly superficial in argument, virtually void for technical and scientific contribution, with contents largely disconnected from the authors’ claimed intent.
Author Response
Please see the attached for your review and comments.
Author Response File: Author Response.pdf
Round 2
Reviewer 3 Report
I was asked to give an opinion on the revised manuscript. My recommendation after the first review was: reject, on the grounds that I do believe that the defects found in the submission can only be mended by a clean-slate approach, which questions the initial premises without being lenient on the original structure, approach and narrative.
The authors have decided to resubmit, and, as I feared, made numerous and somewhat extensive in-place additions which take for granted that what was left untouched was fine.
My view, instead, is that such changes are not sufficient to remedy the defects that led me to recommend outright reject for this work.
Keeping the same structure (for flow of discourse and causal chain) retains the severe defects noted in my previous review comments. Adding text is not the solution, which can only come from rethinking the intent, the approach, and the claims.
Author Response
Please see the attached for your review and comments.
Author Response File: Author Response.pdf
