Prediction modeling algorithm, including ANN, regression analysis models and other are devoted to identified descriptive models, that could be a component part of an optimization model. Optimization algorithms, including heuristic ones are devoted to find a solution (or a population of them, for conciliating solutions with other linked tasks). For optimizing it is a complete theory (in developing process) that take into account the type of the models that describe the objectives and the restricted functions. Depending of the situation the optimization task coul be mono or multiple objetive, mono or multiple level. I hope your question was ansered according you necessity.

Dear Karun Malavya,

Generally, there are many predictive analysis models such as in classification we can lead predicting a value that is discrete through the category and finite with no order, or in regression we make predicting a value that is continuous through numeric quantity and infinite with ordering.

An optimization algorithm is a procedure which is executed iteratively by comparing various solutions till an optimum or a satisfactory solution is found. With the advent of computers, optimization has become a part of computer-aided design activities.

For more information about this subject i suggest you to see links and attached files on topic.

https://www.amazon.com/Automating-Design-Data-Mining-Algorithms/dp/3642025404

Article Algorithm Runtime Prediction: Methods & Evaluation

Book Handbook of Research on Emergent Applications of Optimizatio...

Best regards

Modelling is different from Optimization. Modelling is the process that predict any response characteristics for a particular set of process parameters.

Optimization uses model equations to maximize/minimize any response characteristics.

Modelling and optimization use: Both Statistical (Taguchi/RSM) as well as soft computing approaches (ANN/Fuzzy logic for Modelling, GA/ACO/ABC/Simulated Annealing/TLBO for Optimization).

Dear José Arzola-Ruiz, Aparna Sathya Murthy and Mohamed-Mourad Lafifi,

Thank you so much for your kind reply.

Dear Dr. M. Chandrasekaran ,

Thank you for your effort to answer the question completely.

Optimization implies changing the system model parameters to produce the 'best' output for a given set of inputs. "Best" can be measured using varying metrics : resources, time, accuracy etc. Prediction implies using the known model parameters to estimate system output behaviour at some specified input state. Both processes can use statistical and probabilistic methods.

Optimization is obtaining the best out of available resources. It will find the optimal conditions of your input parameter combination to achieve our objective of maximizing or minimizing.

But on the other hand, predictive modelling approach develops a model based on your input and output and predicts the output for a set of input conditions such as regression models, ANN and ANFIS.

I suggest you to follow https://www.osti.gov/servlets/purl/206444

https://s3.amazonaws.com/academia.edu.documents/40595736/EAAI10.pdf?response-content-disposition=inline%3B%20filename%3DA_comparative_study_of_population-based.pdf&X-Amz-Algorithm=AWS4-HMAC-SHA256&X-Amz-Credential=ASIATUSBJ6BAOQTLSXW3%2F20200531%2Fus-east-1%2Fs3%2Faws4_request&X-Amz-Date=20200531T080727Z&X-Amz-Expires=3600&X-Amz-SignedHeaders=host&X-Amz-Security-Token=IQoJb3JpZ2luX2VjEL%2F%2F%2F%2F%2F%2F%2F%2F%2F%2F%2FwEaCXVzLWVhc3QtMSJHMEUCIQDcYk4%2BUugG2t0BsRxX%2Bza1XTPF1l3JIYokVt2Dh%2F7dfgIgUdYzQ9Lml4j8Ihd8hqs%2BzFFHx0Pk2AKRFZL8eYGdHiYqtAMIKBAAGgwyNTAzMTg4MTEyMDAiDM3cZJL47HLzGPC%2BNSqRAww4s9QET8P3vpyUk9fsX%2B26NRdRZHQ8R%2F%2FbogwOI4bbhMYrWm1GVcBw4SXEQKSmsObL5MCPrG1J45tG5GhdJewgrWgfjqHg9KKaaKo%2FZr2cC%2BxYxtF%2BTyRqtOML2Hsy2BhR3s85QemyVe3a6JA20eTEYedh%2FCckY979nlaS2epQfPfkxtIYVhAJxc7MWADKhEmUTNy%2BhJUIFEmgfWj7CZ6usBxQMFJo%2FSKaSX0GM4NV07tVF%2BHsPFAepyLawl%2F8HijS2KPI2LT3Py5Rkjo2gvjUSioBRVKa00p4UaRr2X2ib%2FVwr%2BoATUBwKdx6RL51zwrrod9RPm96a%2Byx1y9x9KiFHl0eyVymyoIbFfTkibYGAttHGxk5mRIuwYkbZiwyaq9q0SFnWJZkAowWnzzfx7jUmlUsCdGNWVL2AeLRhB2VkCqPDIotd3xSnELS4yk4t94uLaxA7ZeaAnZDyWfzX8VMCivBtli%2B1FIjA%2FI%2FqGOF7Cf1NLHXMLee7w58ozAcK58VFimrPpHR9fvGSPul%2BQUsMMmtzfYFOusBWxpr6Ul3tRUAhV3VSD5lMde5ePOv5tH9d0fQtgb5d1i7KINkuBixqqU1drhlyB6YIPB6jEmSvKnjSurrd%2BrI1PJvm1S7cY93qAdMgcRMMnScT9Bewk1%2BBmB0%2BH9vsUElfDRdfgeWES1xvX3VBy657%2BpBYlN2U5sCYrYBvIpPc4s8nT6I3xtGnWCYW2KtA3j2NoNVy2lxg4YRuuI%2BQuHLs1BZcSpWVKM5JF%2B3anKuLhc40%2FlrbVpK1dZqbX%2FvD9DE1yhOZgGXQkR4F6T2H87qcSWe1MtTunqafKC55iT9CP5rWxl30zgXMxFtrw%3D%3D&X-Amz-Signature=d10d690964c55f65effd48dff33a0161ea779a72dc7ae56aa88dc588b71b122a

As far as the basics are considered they seem to be the same , basic optimization problems look for dependent and independent variables , in that sense they are similar.

Interesting discussion

You are mixing concepts.

An optimization model consist of a set of objective and restricted functions. For solving this kind of models a wide set of methods exists including heuristics such as Genetic Algorithms and Particle Swarm Optimization methods, but also exists a group of other possible exact and heuristics methods. Wich of them is better depends on the characteristics of the model.

The model itself is the result of the systems analysis of the studied problem. In my researchgate profile you can find enought approaches for solving your problem .

The objective and restricted functions could be described using a set of methods. The first of them could be described in the literature related to your problem. In the case such desciptions are not, then you must to find its. Depending on the characteristics of the relations you could be obligated to organize an experimental plan and to find descriptions using regression analyisis, ANN, Taguchi method and other. But your description could be imprecise. In this last case you must to apply Fuzzy component to your functions .

Also exist Response Surface Methodology (RSM) that are used for finding experimetal optimal solutions. I do not like this method, but you could to use it.

I hope my suggestions will be useful for you. I am ready to answer to any your question.

Pl refer our Journal paper titled "Application of soft computing techniques in machiningperformance prediction and optimization: a literature review", Int J Adv Manuf Technol (2010) 46:445–464

DOI 10.1007/s00170-009-2104-x