Introduction
In the recent past, it has become very important for electricity suppliers all over the world to estimate their future demand. In the past twenty years, analysts have developed various tools to help in forecasting demand for electricity utilities. In the 1970s and early 1980s, it was possible to estimate how much electrical power will be required in the future by simple extrapolation of past energy consumption data. The power utility companies used demand for power that was recorded in the past to determine future demand for the commodity. However, various factors in contemporary global market have made it necessary to adjust the modeling techniques used by power utility companies.
The factors necessitating the change in the mindset of the demand analysts include, among others, the rising cost of fuel, new technologies of producing electricity, high inflation rates (both in the developing and developed economies), and change in lifestyle and institutional structures. As a result of these factors, the power utility companies find it important to take into consideration prices, population dynamics, income, technology, as well as other demographic, economic, technological, and policy variables (Industry Forecast Scenario, 2011).
According to Energy Report (2011a), precision is the most important aspect in the process of forecasting demand. In the past, throughout the world, whenever the demand of electricity was underestimated, it was corrected by setting up generator plants that ran on gas or oil. The setting up of such plants required only a small investment and it was possible to complete the setup within a relatively short period of time. Likewise, overestimations were corrected as the demand for electricity increased with time.
Demand forecasting was made easy given that most countries did not have the environmental concerns that exist today. What this meant was that regulation was more predictable then than it is today. Today, underestimates will eventually end in under-capacity, which will lead to brownouts and blackouts. Overestimation, on the other hand, would lead to creation of plants that will remain unused or underutilized for many years. Such a development will be very costly for the utility (Energy Report, 2011a). Furthermore, in light of the increasing role of players from the private sector (as well as the ongoing reforms in the UAE that require demand to be assessed realistically), electricity demand forecasting makes a lot of sense. The current paper aims at outlining how the Abu Dhabi Water and Electricity Company (herein referred to as ADWEC) forecast the demand for its services.
Company Background
The Abu Dhabi Water and Electricity Company is one of the five subsidiaries of the Abu Dhabi Water and Electricity Authority. Abu Dhabi Water and Electricity Company is the only organization that buys and sells water and electricity in Abu Dhabi. The main role of the company is to make sure that all electricity and water consumers in Abu Dhabi have constant supply of the commodities. According to Adwec.ae (2012a), “this role (of providing a constant supply of water and electricity to consumers) is fulfilled through a careful short– term and long– term balancing (act) of supply and demand (Demand Forecast and Generation Expansion Planning)” (par. 5). Adwec.ae (2012a) continues to state that the balancing is achieved through “long– term Power and Water Purchase Agreements (herein referred to as PWPAs) with the Power & Water Producers, through the Bulk Supply Tariff (herein referred to as BST) sales agreements with the Distribution Companies (DISCOs) and Fuel Supply Agreements (herein referred to as FSA) with Fuel Supplies” (par. 5). The company is certified under OHSAS 18001:2007, ISO 9001:2008, and ISO 14001:2004 (Adwec.ae, 2012b).
On July, 2012, the company managed to generate power beyond the 10,000 MW mark. Eighty percent of the power so generated was used in Abu Dhabi, while the rest was sold to other Emirates. What this means is that the company had the capacity to generate electricity that satisfied the needs of the local consumers, exporting the rest to other regions. In the GCC region, only Kuwait and Saudi Arabia have managed to generate more than 10,000 MW of electrical power. According to economic estimates, the company is expected to continue generating power at this level for the remaining part of the 2012. The major reason why this is so is because of the increasing level of export to the Federal Electricity & Water Authority (herein referred to as FEWA) and the Sharjah Electricity & Water (herein referred to as SEWA) in the north. The company is headed by Abdulla Saif Al Nuaimi, who acts as the Chairman. Al Nuami is assisted by three board members (Adwec.ae, 2012a). Figure 1 below is an illustration of the organizational structure at ADWEC:
Results and Discussion
Today, power companies all over the world have realized the need to determine the demand for their products in the future. To this end, there are various methods that power companies use to estimate the demand of electrical power. Companies choose between the different methods depending on the data they have and the detail and nature of the forecasts to be made. ADWEC have realized the need to make accurate predictions of demand for electricity and water in Abu Dhabi and the other emirates it exports to. To attain these accurate predictions, ADWEC uses various methods and compares the different findings made. The approach of using different methods has come in handy for ADWEC, especially for the purposes of setting tariffs and designing demand management programs (Industry Forecast Scenario, 2011).
Trend Method
It is one of the various methods used in predicting demand for electricity. In this method, the variables that are needed are expressed as functions of time. The variables used are not related to technological, policy, demographic, or economic factors. The time function provides adequate information to explain available data, which helps in making short- term projections (Energy Report, 2011b).
ADWEC uses the trend method to predict the demand for power by most consumers, with the exception of high temperature industries. The company makes necessary adjustments to the method to detail unmet demands that arise from power cuts (Industry Forecast Scenario, 2011). According to Adwec.ae (2012c), the trend method is simple and easy to use. However, it has one major shortcoming that negatively affects its effectiveness. The method does not take into account the correlation between the variable under focus and external economic factors.
For instance, it overlooks the impacts of such factors as level of income, growth of cities, population dynamics, as well as costs on the demand for power. The main assumption of the method is that time determines the value attached to the variable under examination. In other words, the method assumes that the variable will maintain the pattern recorded in the past. In spite of these weaknesses, the method is still widely used in predicting the demand for electricity. The major reason why it is still used is the fact that it helps the company to generate an estimate of the value that is predicted (Industry Forecast Scenario, 2011).
End-Use Method
In addition to the trend method analyzed above, ADWEC uses the end-use method predict demand. The technique takes into consideration the effects of the patterns that result from using electricity on different systems and devices. The technique is used to estimate demand in agricultural, commercial, industrial, and residential sectors. Like the trend method, this method makes several assumptions in predicting demand for power. The main assumption made in this technique is that electrical power is not needed as the final good. On the contrary, the power is needed for service delivery (Industry Forecast Scenario, 2011).
The equation below shows how energy is consumed in different sectors:
E=N*S*P*H
Where,
- E= Energy consumed by an appliance (kWh)
- N= number of customers
- S= the number of such appliances for each customer
- P = power needed to operate the appliance (kWh)
- H= hours the appliance is used
The summation of E for various end-uses in the sector provides an estimate of how much electrical energy is used in the sector. The P value in the equation accounts for energy use efficiency, inter-fuel substitution, and rates of utilization. In this manner, the method implicitly factors in income, prices, as well as other policy and economic effects.
At times, ADWEC combines the end-use method and the trend method to estimate the amount of energy that is needed in diverse sectors in the future (Energy Report, 2011a). The technique is crucial in cases where new fuels and technologies are brought into the market. The technique is also useful in the absence of time series data on consumption trends. The method is, however, termed as a mechanical technique, which does not factor in the behaviors of consumers. In addition, it ignores the changes in consumption that are caused by cultural, socio-economic, and demographic factors.
Econometric Method
It is another strategy used in predicting demand. It brings together statistical approaches and economic conceptualizations to come up with prediction equations. ADWEC uses the equations so generated to forecast the demand for electrical power in the market. The method is effective in establishing causal relationships between the demand of power and economic variables. The technique achieves this by using ‘pooled’ cross-sectional data or time-series (Energy Report, 2011b). The demand for electricity is conceptualized as an aspect of different factors. The factors in this case include, among others, population, price of power, and the price of alternative fuel. By putting into consideration all these factors, the technique generates an equation for predicting the demand for electricity. The equation generated is given below:
ED = f (T, POP, Pj, Pi, Y)
Where,
- ED = Electricity Demand
- T = Technology
- POP = Population
- Pj = Price of fuels
- Pi = Own Price
- Y = Income or output
The company uses a variant of the econometric method, where it applies a 4.2% annual population growth rate to estimate the demand for power for a period of fifteen years. The method requires consistent data collected over a long period of time to determine how variables relate to each other in the short term and the long term. However, the method is criticized for assuming that the rate of growth for explanatory variables will remain constant. Moreover, the method does not factor in economic shocks and policy measures that may change how the variable in focus will behave in the future.
Time Series
Time series techniques are econometric methods that use the variable’s lagged values as explanatory variables. The main assumption of the method is that in the future, variables will behave in a manner related to their predicted and actual values in the past (Industry Forecast Scenario, 2011). The method puts into consideration adjustments that factor in how past realizations deviated from what was expected. In this regard, for a time series technique to be effectively used in predicting future demand for electricity, the data for twenty to thirty time periods is needed. Various characteristics set apart econometric methods and time series techniques. For example, the former relies on time series data. Another character setting them apart is the nature of explanatory variables taken into consideration. In the former, such explanatory variables as income, population growth, and prices are used to determine the causal relationship. However, in the case of time series methods, only the variable’s lagged or previous values are used to make predictions (Energy Report, 2011a).
ADWEC uses time series techniques to estimate the demand for electricity in the short-term period. For example, the company has a time series model that is used to determine the monthly demand for power for a period of three years. On the other hand, the company uses econometric methods to determine the demand for electricity in the long term (Energy Report, 2011b). Time series techniques have the merit of structural simplicity compared to the other techniques used in predicting demand for electricity. Furthermore, in series techniques, there is no need to collect the data of multiple variables.
All that is required to use the technique is the ability to come up with accurate observations of the particular variable that is under examination. However, the techniques have various shortcomings that negatively affect their effectiveness in predicting the demand for electricity. The major weakness of these models is that they fail to establish the nature of the cause and effect link between the various variables. In the case of ADWEC, the model does not provide information as to why there were changes in the demand for electricity, which is the variable under study (Industry Forecast Scenario, 2011).
Conclusion
ADWEC has grown to become one of the leading generators and suppliers of electricity in the United Arab Emirates and in the larger GCC region. The main reason for the success of the company is the ability to forecast how demand for electricity changes with time. The company uses different methods to forecast demand, depending on whether the forecast is short term or long term. It also takes into consideration the needs of each sector of the economy.
References
Adwec.ae. (2012). ADWEC organizational structure. Web.
Adwec.ae. (2012). ADWEC’s role. Web.
Adwec.ae. (2012). ADWEA’s peak electricity generation exceeds 10,000 MW. Web.
Energy Report. (2011). Energy industry report. UAE Energy Industry Report, 3(1), 6-12.
Energy Report. (2011). Energy industry report. UAE Energy Industry Report, 3(2), 3-9.
Industry Forecast Scenario. (2011). Energy prediction in UAE. UAE Oil & Gas Report, 2(2), 39-47.