Science topic

# Demand Analysis - Science topic

Explore the latest questions and answers in Demand Analysis, and find Demand Analysis experts.

Questions related to Demand Analysis

I am trying to predict peak demand using machine learning techniques. Current articles consider this as a time series prediction issue and consider a 7-day lag to predict peak demand. A ML model I am trying to apply considers new features for this prediction, and I applied it without a week prior value lag. I was challenged why I did not use lag values for time series prediction like this issue.

The objective of my project was to evaluate whether adding new features would improve the daily peak demand prediction and assess the effects of the new features. If I use new features to predict daily demand, should I also consider the previous seven days' lags as a new feature? Is it correct to combine several COVID-19 related features with the lag demand for peak demand prediction for an unstable situation like COVID-19?

Ps:

1- The model I used for prediction is LightGradient Boosting.

2- Data trained and tested during COVID-19 situation (2020 & 2021)

3- The weekly trends of my target value in 2020 and 2021 are as below figures.

Thinking about food crop production this year, will there the surplus or shortage as a result of the global pandemic?

I'm using microdata on a household basis for analyzing demand patterns in Indonesia. By using the Indonesian Family Life Survey (IFLS) I have plenty of good data, but some variables are not available such as price in certain commodities especially non-food data prices. If I get another data from the Central Bureau of Statistics (macro data) to generate prices based on the same location and year. Could it possible to merge both data as one and run estimation? All I need is price data because IFLS didn't provide enough data for non-food prices.

Good afternoon,

My question is on what are the recent advancements and improvements in the estimation of ductility demand and behaviour factor relations? Of course we have Miranda and Bertero (1994), we have Priestley, Calvi and Kowalski book on DDBD etc, but what are the new findings in the last 5 years?

Hello Everyone

I am trying to combine two online data sets together (the electricity prices and electricity consumption/production). the two data sets are changing continuously (giving something similar to the heartbeats).

I want to build a mathematical model so it can predict when its the right decision to buy/sell electricity or charge/storage electricity.

I have some experience in MATLAB, Power BI, excel. but as far as I know, they are all dealing with static databases, not with changeable ones.

Could anyone advise me of the best tools or programs to achieve that?

Thanks in advance

Considering that transport is the main component of the logistics system, making domestic production more competitive is basically a strategic decision.

How can increased demand for a city's major roadway make it more competitive?

In current analysis of DNA the multiplexing and barcoding requires high quality and purified sample.Thus finding a significant kit for isolation and satisfying a downstream application is vital for an experiment.As every lab would work to improve their productivity by using a technique which is easy to handle,non hazardous reagents,simple,efficient and convenient.I would like to explore the best top techniques that satisfies at least maximum of these criteria.

I'm exploring non linear modelling technique for demand forecasting in transportation sector where alternative transport mode implement.

As a start it is good to have some ideas from you

Thank you

I would like to explore diffusion models like those presented by Bass (1969) and Mansfield (1961), founded on epidemic approaches.

Most relevant works discussing Plug-in Hybrid Electric Vehicles (PHEVs) consider the effect of PHEVs solely. I wonder if it should take into account other factors, e.g. air-conditioners or other appliance.

What packages should be used and are there any tutorials geared specifically toward this type of analysis?

I like To use PODS for simulation and evaluation of my forecasting algorithms which I have designed for demand forecasting in Airline industry, but I have not found it yet.

Hello everyone,

I am working on demand response management and I would like to have suggestions of possible work in the same. Also, I would like to have a discussion on the same topic about previous work been done and being done.

Most experimental research has assumed a uniform demand distribution and proved that the average orders differ significantly from the optimum. Let's assume two demand distributions - one a normal distribution with mean as 100 and standard deviation as 25, and other a uniform distribution varying between 25 and 175 units. Both distributions have a mean of 100 units. If we assume a 3 sigma variation for the normal distribution, both distributions would have a similar range.

A critical fractile of say 0.75 would mean an order of 117 units (difference of 17 from mean) in case of the normal distribution and it would mean an order of 138 units (difference of 38 from mean). Assuming participants anchor close to the mean, it would be easier in the case of Uniform distribution to prove significant difference than in case of normal distribution.

Inventory control poses a challenge for petrol stations when demand for petroleum products fluctuates from time to time.

I am in the process of defining and setting a neural network for time series forecasting in demand side load profile forecasting in a rural village microgrid application. I know there are algorithms for moving average MA and autoregressive AR techniques to statistically estimate the optimal number of coefficients in such models.

For artificial neural nets ANN and/or for an recurrent neural nets RNN, - is there a proven way for estimating the optimal number of hidden layers (or the number feedback links in an RNN) in the neural net definition - advice would be highly appreciated ?

We require it for the design of sea water RO plant, the open Arabian sea values are assumed to be lower, from the manometric analyzer method the values are as high as 1200 ppm, but the winkler method is showing very low as 6ppm. Which method should we assume correct and what are typical for sea water.

We are working on an integrated synthesis TrnSys model for a parabolic dish based thermal electric power generation system in a micro combined cooling heating and power (micro CCHP) configuration, with active demand response and interactive energy management and control.

Some smartmeter data have been used to compile certain electricity and energy load profiles and energy load contours for residential houses http://www.smartmeters.vic.gov.au/about-smart-meters/reports-and-consultations/advanced-metering-infrastructure-customer-impacts-study-volume-2/appendix-e-confidence-intervals-around-electricity-usage-profiles

Looking for more real-time dataset (ie .xls or .xlxs or .csv or .ods or .zip time series data or profile sequence data) to evaluate computer simulation models for a solar thermodynamic trigeneration system (combined cycle data or daily household load or user demand pattern data). Require such powerplant or household usage datasets for use in training artificial intelligent scheduling and multi objective control aspects in isolated or rural microgrid and smartgrid.

Would appreciate if anyone could inform of available data for community, residential, shack neighborhood or rural village settlements in Africa, South America (e.g. Brazil, Argentina, etc), China, India. Alternatively any other load cycles for any other area will also be helpful (including smart household data).

I am trying to assess the demand for selected food items using the LA/AIDS model. There are more than 1.0 food items consumed by the household. I would like to find out how the Stone Price index can be calculated using the stata software.

In order to reduce the system overstress at maximum demand periods.

I'm planing to do research about demand of cigarettes. One of my independent variable is price of cigarettes. But I am confused on what price I should use- real or nominal price? Someone please help to explain about it. Thanks in advance.

When we estimate an AIDS model in levels, the sum of shares is equal to '1' and adding-up restriction on intercept is '1'. However, if use the model in first difference, the sum of shares in first difference is '0'. In this case, the adding-up restriction on intercept should be '1' or '0'.

I know about some existing methods like: Mobile Average, Exponential Smoothing, Multiple Regression, etc. What about the most recent-advanced-efficient forecasting technique (if it exists)?

I want to use Pv panels as a renewable resource, and consuming the neighborhood level is a part of smart city. It means there is a smart Grid as well. So DR and DSM solutions make sense here.

There is V2G,machine wash dishes, air conditioner and etc.

Which software do you think to model and also for programming is the best one in this case?