019 2013 Predictive Modeling In Sports Leagues An -Books Pdf

019 2013 Predictive Modeling in Sports Leagues An
04 Dec 2019 | 48 views | 0 downloads | 10 Pages | 395.39 KB

Share Pdf : 019 2013 Predictive Modeling In Sports Leagues An

Download and Preview : 019 2013 Predictive Modeling In Sports Leagues An

Report CopyRight/DMCA Form For : 019 2013 Predictive Modeling In Sports Leagues An


SAS Global Forum 2013 Applications Development, Past studies have shown that pay cannot be adequately explained by past performance alone nor are pay levels. justified by future performance in IPL Dalmia 2010 Further evidence was found by Karnik 2009 that more. expensive players often provide a lower rate of return to the owners which brings into question the validity of bidding. up a player at auction This makes very hard for the franchise to spend money on the players efficiently So the. franchisee owners need a robust model that can be used to estimate the player s bid amount. In this paper we constructed several predictive models to get the probability of selection of each player that may be. used as a valuation factor in the cost function cost function can vary from one franchise to other for estimating his. bid amount Data from the Twenty20 cricket including first four seasons 2008 2011 of the IPL was used as the input. to the model to predict the selection of each player in the fifth season 2012 The rest of this paper is organized as. follows Section 2 provides a brief literature review Section 3 describes the methodology Section 4 presents results. and section 5 concludes the paper by presenting the future work. Team Name Owner s, Mumbai Indians Mukesh Ambani Owner of Reliance Industries. RoyalChallengers Bangalore Vijay Malya UB Group, Hyderabad Deccan Chargers T Venkatram Reddy Deccan Chronicle group. Chennai Super Kings India Cements Gurunath Meiyappan public face. Delhi Daredevils GMR Group, Kings XI Punjab Ness Wadia Preity Zinta Dabur Apeejay Surendera. Kolkata Knight Riders Shahrukh Khan Red Chillies Entertainment Juhi. Chawla Jay Mehta, Rajasthan Royals Emerging Media Lachlan Murdoch Shilpa Shetty Raj.
Pune Warriors introduced in 2011 Subrato Roy Sahara. Kochi TuskersKerela introduced in 2011 and defunct Kochi Cricket Private Ltd. Table 1 IPL Teams and Owners 2008 Current,LITERATURE REVIEW. A variety of studies have looked at performance and franchise bidding in the Indian Premier League This paper is. first ever study on the selection of players in the IPL teams and estimation of their bid amounts in the Twenty20 form. of cricket Following is a brief review of the work done on the IPL and related studies. Parker Burns and Natarajan 2008 explored the determinants of valuations and investigated a number of. hypotheses related to the design of the auction in IPL using information of the previous performance experience and. other characteristics of individual players, Iyer and Sharda 2009 used neural networks in forecasting the selection of athletes in the cricket teams by predicting. their future performance based on past performance A prediction for the selection of a cricketer in the one day. international world cup 2007 was made To predict the selection players were categorized into a performer a. moderate or a failure, Karnik 2009 followed a very simple approach to derive the hedonic price equations for estimating a bid amount for. each cricketer in the Indian Premier League IPL auction He developed price models using the data from the 2008. season and successfully tested against the data from the 2009 season The variables used in the equations were the. common playing factors such as runs scored wickets taken and age He observed a lower rate of return from the. expensive players to the owners of the teams that showed the inefficiency in judging the pay levels of the players by. the bidders, Singh Gupta and V Gupta 2011 formulated an integer programming model for the efficient bidding strategy for the. franchises The model was implemented in a spreadsheet that helped in taking bidding decisions in real time and. overcome winner s curse which is typically associated with normal bidding processes. Singh 2011 made an effort to measure the performance of teams in the IPL using the non parametric mathematical. approach called Data Envelopment Analysis DEA He used both playing and non playing factors for analyzing the. efficiencies of the teams in 2009 season,SAS Global Forum 2013 Applications Development.
Lenten Geerling and K nya 2012 analyzed various playing and non playing factors of the athletes of cricket sport. that determine their bid value in the auction of Indian Premier League They considered every form of cricket to find. the relationship between those factors and the wages of players Several cross sectional models were estimated to. find the combinational effects of the variables on the salary of the players Most of the studies have focused on. making the bidding decisions in the IPL but this paper is the one of the first studies focused on prediction of the. selection of players in the IPL teams and estimation of their bid amounts Due to the unavailability of sufficient data. league started in 2008 other scholars have used the data from other forms of cricket instead of Twenty20 but our. models were built using data only from Twenty20 type of cricket which ensured the relevancy of our models for this. exciting and new form of cricket,Variable Level Variable Description. All Rounder Binary Indicates whether is a player is an all rounder or not. Value is 1 if batsman otherwise 0, Innings Interval Number of innings played by the player instead of matches. Not out Interval Number of times a player has been not out in his career. Runs Interval Total number of runs in the Twenty20 career. HS Interval Highest runs in an innings by the player. Average Interval Total number of runs a player has scored divided by the number. of times he is out,Country Nominal Players country. 12 dummy variables were created for the 13 countries. Strike Rate Interval Average number of runs scored per 100 balls faced. Century Interval Number of hundreds in Twenty20 career. Half Century Interval Number of fifties in his Twenty20 career. 4s Interval Number of fours in all the innings he has played in twenty20 so. 6s Interval Number of sixes in all the innings, Catch Interval Number of catches taken by a player during fielding. Result Target Binary Selected in this IPL or not 1 selected and 0 not selected. Table 2 Batting Dataset,Variable Level Variable Description.
Innings Interval Number of innings played by the player instead of matches. Balls Interval Number of balls bowled,Runs Interval Total number of runs conceded. Wkts Interval Number of wickets taken, Average Interval Average number of runs conceded per wicket. Country Nominal Players country, 12 dummy variables were created for the 13 countries. Economy Interval Average number of runs conceded per over. SR Interval Average number of balls bowled per wicket taken. Best Interval Best innings bowling, 4w Interval Number of innings in which the bowler took at least four wickets. 5w Interval Number of innings in which the bowler took at least five wickets. Result Target Binary Selected in this IPL or not 1 selected and 0 not selected. Table 3 Bowling dataset,SAS Global Forum 2013 Applications Development.
METHODOLOGY, We culled data from cricketarchive com www iplt20 com sports ndtv com and www espncricinfo com Data consisted. of all playing factors excluding fielding and wicket keeping These two abilities also contribute towards the. performance of the players but the scope of this article is to predict the selection of batsmen and bowlers in the. teams We created two datasets one for bowlers and another for batsmen The players who can play a role of. batsman and bowler both known as all rounders and have a different role in the team It would have been better if a. different dataset for all rounders is created but due to the small sample size we included them in batsmen dataset. with a variable named All rounder that differentiated them from batsmen The batsmen dataset contained a total of. 13 independent variables and one dependent binary variable that shows whether a player is selected in the team or. not There were 11 input variables and one target variable in the bowling dataset Twelve dummy variables were. created for thirteen counties in both datasets to convert qualitative facts into numeric values For complete. Figure 1 SEMMA, information about all variables see table 2 and table 3 Predictive modeling had been done in a sequence of steps. known as SEMMA Sample Explore Modify Model and Assess developed by SAS Institute Inc as shown in the. We collected the data containing the information of 313 batsmen and 277 bowlers of different countries who have. either played in any season of IPL or have ever registered themselves for the auction These datasets contained the. information of the players from their Twenty20 career including the four seasons of the IPL from 2008 to 2011 Using. the data prediction of the selection of players in the 2012 season was made Following the SEMMA data mining. approach a sequence of steps was performed on both bowling and batting datasets for predictive modeling as shown. in figure 1 After collecting the data the next step was to create a balanced sample having equal number of 1s and 0s. in the target variable For adjusting the oversampling prior probabilities had been set equal to the percentage of. positive result in the datasets Data was partitioned into 80 training data for the modeling purpose and 20. validation data for the validation purpose using stratified partitioning method. Several exploratory tools such as crosstabs histograms pie charts correlations etc were used to understand the. relationship between target variable and other input variables We found that the variable Country Players country. was strongly associated with the target variable and many input variables in both datasets But this is a structural. relationship that happened to exist in the data due to nature of the team composition rule that a team must contain at. least 14 Indian players See appendix 1 for the team composition rules. To avoid the problem of multicollinearity principal component analysis was done on the input variables Principal. component analysis converts highly correlated variables into a set of linearly uncorrelated variables called principal. SAS Global Forum 2013 Applications Development, components The number of principal components created is less than or equal to the number of original variables. The first principal component captures the largest variance in the data followed by the second principal component. and so on To make our models simpler fewer than all of the principal components were used as inputs as shown in. figure 2 and figure 3 Beyond the selected PC ID yellow line in the figures Eigen values are decreasing rapidly that. show that the principal components with more than PC ID 12 in batting and PC ID 10 in bowling were not able to. capture much variance in the data Twelve principal components were thus selected for the batting dataset that. captured 83 variance and ten principal components selected in the bowling dataset that captured 80 variance in. Figure 2 Batting PCA,Figure 3 Bowling PCA, Output variables from the principal components analysis were used as inputs to the classification models Various. decision tree models regression models and neural network models were constructed to develop the best model that. could predict the selection of a player in the team Various decision tree models with different parameters were. developed for example different combinations of maximum branches maximum depth nominal criterion Chi. square entropy or Gini and significance level for split were used Logistic regression models were constructed with. different combinations of polynomial degree and model selection method Forward backward or stepwise selection. Neural network models were built using combination of different architecture Multilayer perceptron ordinal radial or. normalized radial and number of hidden units, Models developed were compared based on the validation misclassification rate Finally the model that performed.
best was selected in each study and scored against the data containing the actual results of the selection of players. in the 2012 season The selected model in each study provided the probability of selection of each player that could. be used as a valuation factor for a franchisee to make decisions for setting price of each player as described in next. BIDDING FUNCTION, There are various linear and non linear bid functions Gavious Moldovanu and Sela 2002 that can be used by. franchise for bidding in IPL for buying the players Consider n number of owners bidding for an individual player v i is. bidder i s valuation for the player prediction probability from selected predictive model which is private from all other. bidders All bidders other than i perceive vi as a random selection out of the interval 0 1 governed by the. SAS Global Forum 2013 Applications Development, distribution function F and independent of other valuations We assume that F is continuously differentiable and we. denote by f the associated density function We also assume that f v 0 for all v 0 1. A bid x causes a cost g x where g R R is a strictly increasing function The bidder with the highest bid wins the. player We assume that the cost functions are linear g x x Let there be n bidders face linear cost functions then. the bid function of every bidder is given by,b v v Fn 1 v 0 Fn 1 y dy. Some of the important findings from exploratory analysis in both studies are presented below. Main findings in the batting data, 1 Few players who have played more than 120 matches are all selected indicating all experienced players are. included in the squads, 2 This form of cricket encourages youngsters A minimum of 6 players from the BCCI under 22 pool in each.
squad thus most of the selected batsmen have played less than 30 matches. 3 More than 70 of selected players have 50 or more as their highest score. 4 As per the rules for team composition Minimum of 14 Indian players must be included in each squad maximum. players should belong to India and thus most of the selected players are Indians. 5 Batting average is also an important variable and thus most of the players selected have an average more than. Main findings in the bowling data, 1 Most of the Bowlers selected have a best bowling figure of 3 wickets per game or more. 2 Economy rate is a very important factor in all forms of cricket and here most of the selected players have. economy rate between 6 6 and 7 8 which is quite high. 3 More than 90 of selected bowlers played in the last season 2011. 4 Most of the selected bowlers have strike rate of around 20. 5 Most of the selected bowlers have an average between 18 and 30. Franchises bid on a player hoping to buy him with minimum possible spending The selection takes place from a pool. of players consisted of batsmen bowlers and all rounders For the prediction of players several classification models. were developed as explained in the methodology section 80 of the data was used for training the models and 20. for the validation First of all modeling was done using raw variables excluding the highly correlated variable named. Country Some of the important variables found by the top models are presented in the table 4 It can be observed. that 6s Sixes and Average were two most important variables in batting studies Innings and SR Strike rate were. mostly used by the models in the bowling studies, Model Important Variables in Batting Important Variables in Bowling. Stepwise polynomial regression Average 6s Innings SR and Average Runs. Stepwise logistic regression 6s Half Century Innings. Entropy decision tree Runs Not out 6s Innings SR 4w. Chi square tree Average 6s Innings,Table 4 Variable Importance. Due to the high correlation of many input variables principal component analysis transformation was performed in. both studies There was a substantial improvement in the results when new variables principal components instead. of raw variables were used as input to the models, Table 5 shows the results of the top 5 classification models developed for the batting study with raw variables and. with principal components There is a huge difference in the results of two modeling techniques For the batting the. entropy decision tree model resulted in maximum overall correct rate of 87 3 in validation data which shows that it. SAS Global Forum 2013 Applications Development, was a high performance classification model Splitting rules of the entropy decision tree were based on PC 3 being.
most important followed by PC 1 PC 7 and finally PC 4 It is interesting that the multi perceptron MLP neural. network model with three hidden units performed best among all in training data but was not able to give the best. results in the validation data It could not perform well in the validation data perhaps because it was over trained in. the training data The same results can be seen in the ROC curve presented in the figure 4 where the neural network. curve lays on the top of all for most of the time in the training but the entropy decision tree model wins the race during. the validation, Model Validation Misclassification Rate Validation Misclassification Rate. with Raw Variables with PCA,Entropy decision tree 0 429 0 127. Neural network 0 524 0 174,Stepwise polynomial regression 0 460 0 190. Chi square tree 0 429 0 206,Stepwise logistic regression 0 460 0 206. Table 5 Batting Modeling Result,Figure 4 Batting ROC Curve.
Table 6 shows the results of the models built using raw variables and principal components for the bowling dataset. The models constructed with the principal components performed much better than the models with the raw. variables The stepwise polynomial regression model with degree 2 including the main effects outperformed the. classification models in the validation data by predicting the target variable 81 8 correctly as described in table 6. This model was built using 5 principal components PC 1 PC 2 PC 1 PC 2 PC 2 PC 2 PC 2 PC 3 Again it. was noticed that neural network worked best among all for training data but could not do well with validation because. of over training It is on the top of all in the ROC curve in the training data but this time stepwise polynomial. regression model did well in validation data based on misclassification rate as shown in figure 5. Model Validation Misclassification Validation Misclassification. Rate with Raw Variables Rate with PCA,Stepwise polynomial regression 0 473 0 182. Stepwise logistic regression 0 455 0 200,Entropy decision tree 0 455 0 200. Chi square tree 0 473 0 273,Neural network 0 509 0 367. Table 6 Bowling Modeling Result,SAS Global Forum 2013 Applications Development. Figure 5 Bowling ROC curve, The model selected based on minimum validation misclassification rate in each study was scored against their.
respective data containing the actual results Table 7 shows the statistics of performance of the entropy decision tree. model selected for predicting selection of the batsmen There were a total of 159 players selected in IPL and our. model was able to predict 152 players correctly out of 159 i e 95 6 which is really very high Low value of false. positive rate or 1 Specificity i e 24 03 high overall correct rate of 85 9 indicates that the developed model is. Prediction,1 152 37 189,0 7 117 124,Total 159 154 313. Sensitivity 152 159 95 60 Specificity 117 154 75 97 Overall correct rate. 152 117 313 85 9,Table 7 Batting Scoring Results, Table 8 presents the results of the scoring of bowling population with the selected stepwise polynomial regression. model based on minimum validation misclassification rate The sensitivity of the model is 76 55 i e our model was. able to predict 76 55 selected bowlers correctly which is a reasonable number 1 Specificity of this model is only. 16 66 and overall correct rate of 79 8 that indicates the high class performance of the model Selected model in. each study performed extraordinarily well that proves that the predictive models developed could be used to help the. franchise in making efficient selection decisions, Scoring the data with best model in each study provided us the probability of selection of each player that can be. effectively used as valuation factor in the bidding function as described in the methodology section. Prediction,1 111 22 133,0 34 110 144,Total 145 132 277. Sensitivity 111 145 76 55 Specificity 110 132 83 33 Overall correct rate. 111 110 277 79 8,Table 8 Bowling scoring results,SAS Global Forum 2013 Applications Development.
DISCUSSION AND FUTURE WORK, The results show that classification models can perform reasonably well in predicting the selection of players based. on their past performance in the future seasons of Indian Premier League Several predictive models were developed. for the selection of batsmen and bowlers and based on minimum validation misclassification rate the top most model. was selected in each study It was found that use of principal components instead of raw variables makes our models. much efficient and robust For the batting and bowling studies the neural network model performed best for the. training data with maximum overall correct rate but perhaps it was over trained and could not produce best results in. the validation data In the batting studies the entropy decision tree model outperformed the other models by. providing more than 87 overall correct rate in validation data Finally it was able to predict the actual selection of the. players in 2012 season 85 9 For the bowling data the stepwise polynomial regression model with degree 2 did a. good job in predicting the selection in validation data with only 18 2 misclassification rate and predicted the actual. selection in 2012 season 79 8 correctly The performance of the models was quite high which indicates that these. can help decision makers during auction By using our models decision makers can reduce their list of players and. thus make efficient selection decisions The models developed provide a probability measure of selection of each. player in the team We suggest using the probability measure as a valuation factor in the bidding equation to set. salaries for the players as described in methodology section It is also highly recommended for the team selectors in. other sports as well to use principal component analysis when highly correlated variables are present in the data. This study is the first attempt to develop predictive models for the selection of players in Twenty20 form of cricket. Iyer and Sharda 2009 did a similar research and found that the neural network model can predict the selection of. players in the One day international form of the cricket with the overall correct rate of 70 As compared to their. model our models are doing better with more than 79 overall correct rate but in our study form of the cricket is. Twenty20 that requires different skill set Beyond the accuracy of in predicting the selection of players in the IPL. these types of models can also be used to predict the selection in other forms of crickets or even in other sports. This study being an exploratory project has some shortcomings We included all rounders in the batting dataset but. results could have become more efficient if different studies were performed separately on them So we can say that. our models can better predict the selection of only batsmen and bowlers Fielding is another important aspect of. cricket that was not considered as input in our study There are some team composition rules set by the Board of. cricket council of India for the IPL that were also not taken into consideration Using Country variable forcefully as an. input variable was an attempt to follow a rule that minimum of 14 Indian players must be included in each squad but. specific rules were not implemented There may be some non playing factors like age bid cap and other parameters. that are not directly related to the playing abilities of the players that can influence the selection of the players were. also not included in the modeling So the models developed can aid team owners in an objective way to make final. Pay cannot be adequately explained by past performance alone nor are pay levels justified by future performance. Dalmia 2010 So the strategy of using probability of selection of players as a valuation factor in the biding equation. can save franchise a lot of spending on players who are poor performers Our models have the ability to build a. talented team with minimum cost Team selectors can use our models to make better decision on predicting the. performance of the players in future, Future work includes considering several other factors in the analysis like age the ability to lead the team captaincy. money spent on each player in past season bid caps and performance of the players in other forms of cricket. Considering all these factors and team composition rules may result in better models. REFERENCES, Beaudoin D 2003 The best batsmen and bowlers in one day cricket Thesis Canada Laval University. Dalmia K 2010 The Indian Premier League Pay versus performance Thesis New York University. Gavious A Moldovanu B Sela A 2002 Bid costs and endogenous bid caps RAND Journal of Economics. 33 4 709 722, Iyer S R Sharda R 2009 Prediction of athletes performance using neural networks An application in cricket. team selection Expert Systems with Applications 36 3 5510 5522. Karnik A 2009 Valuing cricketers using hedonic price models Journal of Sports Economics 11 4 456 469. Lenten L J A Geerling W K nya L 2012 A hedonic model of player wage determination from the Indian. Premier League auction Further evidence Sport Management Review 15 3 60 71. Parker D Burns P Natarajan H 2008 October Player valuations in the Indian Premier League Frontier. SAS Global Forum 2013 Applications Development, Sharda R Delen D 2006 Predicting box office success of motion pictures with neural networks Expert Systems.
with Applications 30 243 254, Singh S Gupta S Gupta V 2011 Dynamic bidding strategy for players auction in IPL International Journal of. Sports Science and Engineering 05 01 03 16, Singh S 2011 Measuring the performance of teams in the Indian Premier League American Journal of Operations. Research 01 180 184, Zimbalist A S 2002 Competitive balance in sports leagues An introduction Journal of Sports Economics 3 2. APPENDIX 1, Some of the Team composition rules set by BCCI are. 1 Minimum squad strength of 16 players plus one physio and a coach. 2 No more than 11 foreign players in the squad and maximum 4 foreign players should be in the playing eleven. 3 Minimum of 14 Indian players must be included in each squad. 4 A minimum of 6 players from the BCCI under 22 pool in each squad. CONTACT INFORMATION, Your comments and questions are valued and encouraged Contact the author at.
Pankush Kalgotra Oklahoma State University Stillwater OK Email pankush okstate edu. Pankush Kalgotra is a second year graduate student majoring in Management Information Systems at Oklahoma. State University He has two year experience of using SAS tools for Data Mining Texting Mining and Sentiment. Analysis projects He is a SAS certified Predictive Modeler using SAS Enterprise Miner 6 1 In December 2012 he. received his SAS and OSU Data Mining Certificate, Ramesh Sharda Oklahoma State University Stillwater OK Email ramesh sharda okstate edu. Dr Ramesh Sharda is a Regents Professor of Management Science and Information Systems at Oklahoma State. University He is also the Director of the Institute for Research in Information Systems at OSU and the Director of the. Executive PhD in Business Program His research interests are quite wide with the general theme being application. of analytical techniques and information technologies for decision support Besides funded projects and numerous. publications in this broad theme he is a co author of textbooks in Decision Support and Business intelligence areas. and co editor of several book series, Goutam Chakraborty Oklahoma State University Stillwater OK Email goutam chakraborty okstate edu. Dr Goutam Chakraborty is a professor of marketing and founder of SAS and OSU data mining certificate and. SAS and OSU business analytics certificate at Oklahoma State University He has published in many journals such. as Journal of Interactive Marketing Journal of Advertising Research Journal of Advertising Journal of Business. Research etc He has chaired the national conference for direct marketing educators for 2004 and 2005 and co. chaired M2007 data mining conference He has over 25 years of experience in using SAS for data analysis He is. also a Business Knowledge Series instructor for SAS. SAS and all other SAS Institute Inc product or service names are registered trademarks or trademarks of SAS. Institute Inc in the USA and other countries indicates USA registration. Other brand and product names are trademarks of their respective companies.

Related Books

Supplier Counterfeit Awareness Training - Harris

Supplier Counterfeit Awareness Training Harris

SAE Standards AS5553 or AS6174. Warning Flags Price is low or significantly different than price history ... Supplier Counterfeit Awareness Training .




What do we mean by prediction in language comprehension?

What do we mean by prediction in language comprehension

review of the literature led us to the conclusion that different sub?elds and different researchers have criti-cally different conceptions of what it means to predict during language comprehension. This has led to much confusion with researchers sometimes arguing at cross-purposes. The term prediction has become so loaded

Concise Old Testament Survey - tbcokc.org

Concise Old Testament Survey tbcokc org

1 Some of the ideas and the plan used in this Survey for each of the books of the Bible (author and title, date, purpose and theme, key verse(s), etc.) are similar to and drawn from other survey materials such as, Briefing the Bible, J. Vernon McGee, Zondervan, Grand Rapids, 1949; A Popular Survey of the Old Testament, Norman L.

Accomplish Segment Reporting using Document Splitting in ...

Accomplish Segment Reporting using Document Splitting in

Accomplish Segment Reporting using Document Splitting in the New GL by Mitresh Kundalia Director - SAP Practice, Quality Systems & Software (QSandS.com)

OLYMPIAD CORNER - Canadian Mathematical Society

OLYMPIAD CORNER Canadian Mathematical Society

OLYMPIAD CORNER /245 OLYMPIAD CORNER No. 373 The problems featured in this section have appeared in a regional or national mathematical Olympiad. Click here to submit solutions, comments and generalizations to any



ENGLISH USER MANUAL Version 1.0 CLICk HERE to ACCESS tHE tAbLE of CoNtENtS. MANUAL V1.0 tAbLE of CoN tEN S > p. Table Of COnTenTs InTrOduCTIOn ...

San Francisco Bay Area Water Trail Implementation Meeting ...

San Francisco Bay Area Water Trail Implementation Meeting

An agreement was reached recently to merge ABAG and MTC. The merger is scheduled to be completed by July 1. All ABAG staff will be MTC staff as of that date; functions, offices, etc. will not to change. Water Trail staff are not anticipating any major changes to how the Water Trail program is currently operating.

Elastisitas Permintaan & Penawaran

Elastisitas Permintaan amp Penawaran

elastisitas sering tidak menggunakan tanda negatif. ... Alternatif energy untuk bahan bakar mobil (e.g., battery untuk mobil /mobil listrik) mungkin tersedia. D SR. D LR.



4. Obtain the equipment from Ford Motor Company necessary to evacuate the A/C system by submitting a VIN-specific contact via the SSSC web contact site the SSSC web contact site can be accessed through the PTS website using the SSSC link listed at the bottom of the OASIS VIN report screen or listed under the SSSC tab. 5. Once Loaner equipment ...