df <- data.frame(rbind(c('A', '','W','1'),c('A','','W', '3'),c('A', 'BK','R','4'),c('B','BN','T', '12')))

df <- data.frame(rbind(c('A', '','W','0'),c('A','','W', '0'),c('A', 'BK','R','4'),c('B','BN','T', '12')))

# create the sample data frame
df <- data.frame(rbind(c('A', '','W','1'),c('A','','W', '3'),c('A', 'BK','R','4'),c('B','BN','T', '12')))

# add a new level to the 4th column factor
df[, 4] <- factor(df[, 4], levels=c(levels(df[, 4]), '0'))

# write 0 to 4th column where 2nd column is ''
df[df[, 2] == '',4] <- '0'

> df <- data.frame(rbind(c('A', '','W','1'),c('A','','W', '3'),c('A', 'BK','R','4'),c('B','BN','T', '12')))
> df
   X1 X2 X3 X4
1  A     W  1
2  A     W  3
3  A BK  R  4
4  B BN  T 12
> str(df)
'data.frame':   4 obs. of  4 variables:
$ X1: Factor w/ 2 levels "A","B": 1 1 1 2
$ X2: Factor w/ 3 levels "","BK","BN": 1 1 2 3
$ X3: Factor w/ 3 levels "R","T","W": 3 3 1 2
$ X4: Factor w/ 4 levels "1","12","3","4": 1 3 4 2

df$X4<-as.numeric(df$X4)

df[df$X2 == "", "X4"] <- 0

> df
     X1 X2 X3 X4
   1  A     W  0
   2  A     W  0
   3  A BK  R  4
   4  B BN  T  2