# ── Definisi fungsi ──────────────────────────────────────────────────────────
compute_formula <- function(x, formula) {
  valid_formulas <- c("linear", "quadratic", "cubic", "exponential")
  if (!formula %in% valid_formulas) {
    stop(paste("Invalid formula. Choose from:", paste(valid_formulas, collapse = ", ")))
  }
  switch(formula,
    "linear"      = 2 * x + 3,
    "quadratic"   = x^2 - 3 * x + 2,
    "cubic"       = x^3 - 4 * x^2 + x + 6,
    "exponential" = exp(0.3 * x)
  )
}

# ── Hitung semua formula untuk x = 1:20 dengan nested loop ──────────────────
x_values   <- 1:20
formulas   <- c("linear", "quadratic", "cubic", "exponential")
results_df <- data.frame()

for (x in x_values) {
  for (f in formulas) {
    y          <- compute_formula(x, f)
    results_df <- rbind(results_df, data.frame(x = x, formula = f, y = y))
  }
}

# ── Tabel pivot: tiap formula jadi kolom ────────────────────────────────────
results_wide        <- tidyr::pivot_wider(results_df, names_from = formula, values_from = y)
results_wide[, 2:5] <- round(results_wide[, 2:5], 3)

results_wide %>%
  kable(
    caption   = "Tabel 1.1 — Nilai f(x) untuk Setiap Formula (x = 1 sampai 20)",
    col.names = c("x", "Exponential", "Cubic", "Linear", "Quadratic"),
    align     = c("c", "r", "r", "r", "r")
  ) %>%
  kable_styling(
    bootstrap_options = c("striped", "hover", "condensed", "bordered"),
    full_width        = TRUE,
    font_size         = 13
  ) %>%
  column_spec(1, bold = TRUE, color = "white", background = "#424242") %>%
  column_spec(2, color = "#9C27B0", bold = TRUE) %>%
  column_spec(3, color = "#FF5722") %>%
  column_spec(4, color = "#2196F3") %>%
  column_spec(5, color = "#4CAF50")

ggplot(results_df, aes(x = x, y = y, color = formula)) +
  geom_line(linewidth = 1.2) +
  geom_point(size = 2, alpha = 0.7) +
  scale_color_manual(values = c(
    "linear"      = "#2196F3",
    "quadratic"   = "#4CAF50",
    "cubic"       = "#FF5722",
    "exponential" = "#9C27B0"
  )) +
  labs(
    title    = "Comparison of Mathematical Formulas (x = 1 to 20)",
    subtitle = "Linear, Quadratic, Cubic, and Exponential",
    x        = "x",
    y        = "f(x)",
    color    = "Formula"
  ) +
  theme_minimal(base_size = 13) +
  theme(legend.position = "bottom")

sales_df <- read.csv("sales_data.csv")
cat("Dimensi data:", nrow(sales_df), "baris,", ncol(sales_df), "kolom\n")

head(sales_df, 10) %>%
  kable(
    caption     = "Tabel 2.1 — Preview sales_data.csv (10 baris pertama)",
    col.names   = c("Sales ID", "Day", "Sales Amount", "Discount Rate"),
    align       = c("c", "c", "r", "r"),
    format.args = list(big.mark = ",")
  ) %>%
  kable_styling(
    bootstrap_options = c("striped", "hover", "condensed", "bordered"),
    full_width        = FALSE,
    position          = "center",
    font_size         = 13
  ) %>%
  column_spec(1, bold = TRUE, color = "white", background = "#1565C0") %>%
  column_spec(3, color = "#1B5E20", bold = TRUE) %>%
  column_spec(4, color = "#E65100")

# ── Nested function: hitung net sales & kumulatif per salesperson ────────────
apply_discount <- function(amount, rate) {
  amount * (1 - rate)
}

get_cumulative_sales <- function(df, salesperson_id) {
  sub_df                <- df[df$sales_id == salesperson_id, ]
  sub_df                <- sub_df[order(sub_df$day), ]
  sub_df$net_sales      <- round(mapply(apply_discount,
                                        sub_df$sales_amount,
                                        sub_df$discount_rate), 2)
  sub_df$cumulative_net <- cumsum(sub_df$net_sales)
  return(sub_df)
}

simulate_sales <- function(df) {
  all_cumulative  <- data.frame()
  salesperson_ids <- unique(df$sales_id)
  for (sid in salesperson_ids) {
    all_cumulative <- rbind(all_cumulative, get_cumulative_sales(df, sid))
  }
  return(all_cumulative)
}

# ── Jalankan simulasi ────────────────────────────────────────────────────────
sim_result <- simulate_sales(sales_df)

head(sim_result, 10) %>%
  kable(
    caption     = "Tabel 2.2 — Hasil Simulasi: Net Sales & Cumulative Net (10 baris pertama)",
    col.names   = c("Sales ID", "Day", "Sales Amount", "Discount Rate",
                    "Net Sales", "Cumulative Net"),
    align       = c("c", "c", "r", "r", "r", "r"),
    digits      = 2,
    format.args = list(big.mark = ",")
  ) %>%
  kable_styling(
    bootstrap_options = c("striped", "hover", "condensed", "bordered"),
    full_width        = FALSE,
    position          = "center",
    font_size         = 13
  ) %>%
  column_spec(1, bold = TRUE, color = "white", background = "#1565C0") %>%
  column_spec(5, color = "#1B5E20", bold = TRUE) %>%
  column_spec(6, color = "#6A1B9A", bold = TRUE)

summary_sales <- sim_result %>%
  group_by(sales_id) %>%
  summarise(
    Total_Gross  = round(sum(sales_amount), 2),
    Total_Net    = round(sum(net_sales), 2),
    Avg_Discount = round(mean(discount_rate) * 100, 1),
    Max_Net_Day  = round(max(net_sales), 2),
    .groups      = "drop"
  )

summary_sales %>%
  kable(
    caption     = "Tabel 2.3 — Ringkasan Penjualan per Salesperson",
    col.names   = c("Sales ID", "Total Gross (IDR)", "Total Net (IDR)",
                    "Avg Discount (%)", "Max Net/Day"),
    align       = c("c", "r", "r", "c", "r"),
    format.args = list(big.mark = ",")
  ) %>%
  kable_styling(
    bootstrap_options = c("striped", "hover", "bordered"),
    full_width        = FALSE,
    position          = "center",
    font_size         = 13
  ) %>%
  column_spec(1, bold = TRUE, color = "white", background = "#1565C0") %>%
  column_spec(2, color = "#424242") %>%
  column_spec(3, bold = TRUE, color = "#1B5E20") %>%
  column_spec(4, color = "#E65100")

ggplot(sim_result, aes(x = day, y = cumulative_net,
                       color = factor(sales_id), group = sales_id)) +
  geom_line(linewidth = 1.1) +
  scale_color_brewer(palette = "Set1", name = "Salesperson ID") +
  labs(
    title    = "Cumulative Net Sales per Salesperson (30 Days)",
    subtitle = "After applying conditional discount rates",
    x        = "Day",
    y        = "Cumulative Net Sales (IDR)"
  ) +
  theme_minimal(base_size = 13) +
  theme(legend.position = "bottom")

# ── Fungsi kategorisasi performa ─────────────────────────────────────────────
categorize_performance <- function(sales_amount) {
  categories <- c()
  for (amt in sales_amount) {
    cat_label <- if (amt >= 1800) {
      "Excellent"
    } else if (amt >= 1400) {
      "Very Good"
    } else if (amt >= 900) {
      "Good"
    } else if (amt >= 400) {
      "Average"
    } else {
      "Poor"
    }
    categories <- c(categories, cat_label)
  }
  return(categories)
}

# ── Terapkan ke data & hitung persentase ─────────────────────────────────────
sales_df$performance_cat <- categorize_performance(sales_df$sales_amount)

cat_order   <- c("Excellent", "Very Good", "Good", "Average", "Poor")
cat_summary <- sales_df %>%
  group_by(performance_cat) %>%
  summarise(count = n(), .groups = "drop") %>%
  mutate(percentage = round(count / sum(count) * 100, 1)) %>%
  arrange(factor(performance_cat, levels = cat_order))

cat_summary %>%
  kable(
    caption   = "Tabel 3.1 — Distribusi Kategori Performa Penjualan",
    col.names = c("Kategori", "Jumlah", "Persentase (%)"),
    align     = c("l", "c", "c")
  ) %>%
  kable_styling(
    bootstrap_options = c("striped", "hover", "bordered"),
    full_width        = FALSE,
    position          = "center",
    font_size         = 13
  ) %>%
  column_spec(1, bold = TRUE) %>%
  column_spec(2, color = "#1565C0", bold = TRUE) %>%
  column_spec(3, color = "#6A1B9A") %>%
  row_spec(which(cat_summary$performance_cat == "Excellent"), background = "#E8F5E9") %>%
  row_spec(which(cat_summary$performance_cat == "Poor"),      background = "#FFEBEE")

palette_cat <- c(
  "Excellent" = "#1B5E20",
  "Very Good" = "#388E3C",
  "Good"      = "#FBC02D",
  "Average"   = "#F57C00",
  "Poor"      = "#C62828"
)

cat_summary$performance_cat <- factor(cat_summary$performance_cat, levels = cat_order)

p_bar <- ggplot(cat_summary, aes(x = performance_cat, y = count,
                                  fill = performance_cat)) +
  geom_bar(stat = "identity", width = 0.6, show.legend = FALSE) +
  geom_text(aes(label = paste0(count, "\n(", percentage, "%)")),
            vjust = -0.3, size = 3.8, fontface = "bold") +
  scale_fill_manual(values = palette_cat) +
  labs(title = "Sales Performance Distribution — Bar Chart",
       x = "Category", y = "Count") +
  theme_minimal(base_size = 13) +
  ylim(0, max(cat_summary$count) * 1.2)

p_pie <- ggplot(cat_summary, aes(x = "", y = count, fill = performance_cat)) +
  geom_bar(stat = "identity", width = 1, color = "white") +
  coord_polar("y") +
  scale_fill_manual(values = palette_cat, name = "Category") +
  geom_text(aes(label = paste0(percentage, "%")),
            position = position_stack(vjust = 0.5),
            size = 4, color = "white", fontface = "bold") +
  labs(title = "Sales Performance Distribution — Pie Chart") +
  theme_void(base_size = 13)

grid.arrange(p_bar, p_pie, ncol = 2)

company_df <- read.csv("company_data.csv")
cat("Dimensi:", nrow(company_df), "baris,", ncol(company_df), "kolom\n")

head(company_df, 10) %>%
  kable(
    caption     = "Tabel 4.1 — Preview company_data.csv (10 baris pertama)",
    col.names   = c("Company ID", "Employee ID", "Salary", "Department",
                    "Performance Score", "KPI Score"),
    align       = c("c", "c", "r", "l", "r", "r"),
    digits      = 2,
    format.args = list(big.mark = ",")
  ) %>%
  kable_styling(
    bootstrap_options = c("striped", "hover", "condensed", "bordered"),
    full_width        = FALSE,
    position          = "center",
    font_size         = 13
  ) %>%
  column_spec(1, bold = TRUE, color = "white", background = "#0277BD") %>%
  column_spec(4, italic = TRUE, color = "#4A148C") %>%
  column_spec(5, color = "#1B5E20") %>%
  column_spec(6, bold = TRUE, color = "#E65100")

# ── Nested loops + conditional: summary per company ──────────────────────────
generate_company_summary <- function(df) {
  company_ids  <- unique(df$company_id)
  summary_list <- list()

  for (cid in company_ids) {
    sub_df         <- df[df$company_id == cid, ]
    n_emp          <- nrow(sub_df)
    top_performers <- 0

    for (i in 1:n_emp) {
      if (sub_df$KPI_score[i] > 90) {
        top_performers <- top_performers + 1
      }
    }

    summary_list[[length(summary_list) + 1]] <- data.frame(
      Company         = paste("Company", cid),
      N_Employees     = n_emp,
      Avg_Salary      = round(mean(sub_df$salary), 2),
      Avg_Performance = round(mean(sub_df$performance_score), 2),
      Max_KPI         = round(max(sub_df$KPI_score), 2),
      Top_Performers  = top_performers
    )
  }
  return(do.call(rbind, summary_list))
}

company_summary <- generate_company_summary(company_df)

company_summary %>%
  kable(
    caption     = "Tabel 4.2 — Ringkasan per Perusahaan: Gaji, Performa & Top Performers",
    col.names   = c("Perusahaan", "Jml Karyawan", "Avg Salary (IDR)",
                    "Avg Performance", "Max KPI", "Top Performers"),
    align       = c("l", "c", "r", "r", "r", "c"),
    format.args = list(big.mark = ",")
  ) %>%
  kable_styling(
    bootstrap_options = c("striped", "hover", "bordered"),
    full_width        = FALSE,
    position          = "center",
    font_size         = 13
  ) %>%
  column_spec(1, bold = TRUE) %>%
  column_spec(3, color = "#1565C0", bold = TRUE) %>%
  column_spec(5, color = "#E65100", bold = TRUE) %>%
  column_spec(6, bold = TRUE, color = "white",
              background = ifelse(company_summary$Top_Performers >= 3,
                                  "#1B5E20", "#E53935"))

p1 <- ggplot(company_summary, aes(x = Company, y = Avg_Salary, fill = Company)) +
  geom_bar(stat = "identity", width = 0.6, show.legend = FALSE) +
  geom_text(aes(label = scales::comma(Avg_Salary)), vjust = -0.4, size = 3.5) +
  scale_fill_brewer(palette = "Blues", direction = -1) +
  labs(title = "Average Salary per Company", x = "", y = "Avg Salary") +
  theme_minimal(base_size = 13) +
  ylim(0, max(company_summary$Avg_Salary) * 1.15)

p2 <- ggplot(company_summary, aes(x = Company, y = Top_Performers, fill = Company)) +
  geom_bar(stat = "identity", width = 0.6, show.legend = FALSE) +
  geom_text(aes(label = Top_Performers), vjust = -0.4, size = 4, fontface = "bold") +
  scale_fill_brewer(palette = "Oranges", direction = -1) +
  labs(title = "Top Performers (KPI > 90) per Company", x = "", y = "Count") +
  theme_minimal(base_size = 13) +
  ylim(0, max(company_summary$Top_Performers) * 1.2)

grid.arrange(p1, p2, ncol = 2)

# ── Fungsi Monte Carlo ────────────────────────────────────────────────────────
monte_carlo_pi <- function(n_points) {
  set.seed(123)
  x <- runif(n_points, -1, 1)
  y <- runif(n_points, -1, 1)

  inside_circle  <- (x^2 + y^2) <= 1
  pi_estimate    <- 4 * sum(inside_circle) / n_points

  in_subsquare   <- abs(x) <= 0.5 & abs(y) <= 0.5
  prob_subsquare <- sum(in_subsquare) / n_points

  return(list(
    pi_estimate    = pi_estimate,
    prob_subsquare = prob_subsquare,
    x              = x,
    y              = y,
    inside_circle  = inside_circle
  ))
}

# ── Konvergensi dengan berbagai ukuran n ─────────────────────────────────────
n_values   <- c(100, 500, 1000, 5000, 10000)
pi_results <- data.frame()

for (n in n_values) {
  res        <- monte_carlo_pi(n)
  pi_results <- rbind(pi_results, data.frame(
    n_points       = n,
    pi_estimate    = round(res$pi_estimate, 5),
    error          = round(abs(res$pi_estimate - pi), 5),
    prob_subsquare = round(res$prob_subsquare, 4)
  ))
}

cat("Pi aktual:", pi, "\n")

pi_results %>%
  kable(
    caption     = "Tabel 5.1 — Estimasi π Monte Carlo (berbagai ukuran sampel)",
    col.names   = c("N Points", "Estimasi π", "Error vs π Asli", "P(Sub-Square)"),
    align       = c("r", "r", "r", "r"),
    format.args = list(big.mark = ",")
  ) %>%
  kable_styling(
    bootstrap_options = c("striped", "hover", "bordered"),
    full_width        = FALSE,
    position          = "center",
    font_size         = 13
  ) %>%
  column_spec(1, bold = TRUE, color = "#1565C0") %>%
  column_spec(2, bold = TRUE) %>%
  column_spec(3, bold = TRUE,
              color = ifelse(pi_results$error < 0.01, "#1B5E20", "#E65100")) %>%
  row_spec(nrow(pi_results), background = "#E8F5E9")

res_plot <- monte_carlo_pi(2000)
plot_df  <- data.frame(
  x      = res_plot$x,
  y      = res_plot$y,
  status = ifelse(res_plot$inside_circle, "Inside Circle", "Outside Circle")
)

p_scatter <- ggplot(plot_df, aes(x = x, y = y, color = status)) +
  geom_point(alpha = 0.5, size = 0.8) +
  scale_color_manual(values = c("Inside Circle"  = "#1565C0",
                                "Outside Circle" = "#E53935")) +
  coord_fixed() +
  annotate("path",
           x = cos(seq(0, 2 * pi, length.out = 300)),
           y = sin(seq(0, 2 * pi, length.out = 300)),
           color = "black", linewidth = 0.8) +
  labs(
    title    = "Monte Carlo Simulation (n = 2000)",
    subtitle = paste0("pi estimate = ", round(res_plot$pi_estimate, 4)),
    color    = NULL
  ) +
  theme_minimal(base_size = 13) +
  theme(legend.position = "bottom")

p_conv <- ggplot(pi_results, aes(x = n_points, y = pi_estimate)) +
  geom_line(color = "#1565C0", linewidth = 1.2) +
  geom_point(color = "#E53935", size = 3) +
  geom_hline(yintercept = pi, linetype = "dashed", color = "gray40") +
  geom_text(aes(label = round(pi_estimate, 4)), vjust = -0.8, size = 3.5) +
  annotate("text", x = max(pi_results$n_points) * 0.8, y = pi + 0.01,
           label = "True pi", color = "gray40") +
  scale_x_log10(labels = scales::comma) +
  labs(
    title = "Convergence of pi Estimate vs Sample Size",
    x     = "Number of Points (log scale)",
    y     = "pi Estimate"
  ) +
  theme_minimal(base_size = 13)

grid.arrange(p_scatter, p_conv, ncol = 2)

df6          <- read.csv("company_data.csv")
numeric_cols <- c("salary", "performance_score", "KPI_score")
cat("Dimensi:", nrow(df6), "baris,", ncol(df6), "kolom\n")

# ── Min-Max Normalization (loop-based) ───────────────────────────────────────
normalize_columns <- function(df, cols) {
  df_norm <- df
  for (col in cols) {
    mn             <- min(df[[col]], na.rm = TRUE)
    mx             <- max(df[[col]], na.rm = TRUE)
    df_norm[[col]] <- round((df[[col]] - mn) / (mx - mn), 4)
  }
  return(df_norm)
}

# ── Z-Score Standardization (loop-based) ─────────────────────────────────────
z_score <- function(df, cols) {
  df_z <- df
  for (col in cols) {
    mu          <- mean(df[[col]], na.rm = TRUE)
    sigma       <- sd(df[[col]], na.rm = TRUE)
    df_z[[col]] <- round((df[[col]] - mu) / sigma, 4)
  }
  return(df_z)
}

df_norm <- normalize_columns(df6, numeric_cols)
df_z    <- z_score(df6, numeric_cols)

# ── Feature Engineering ───────────────────────────────────────────────────────
df6$performance_category <- categorize_performance(df6$performance_score)
df6$salary_bracket <- cut(
  df6$salary,
  breaks         = c(0, 5000, 8000, 11000, Inf),
  labels         = c("Low", "Medium", "High", "Very High"),
  include.lowest = TRUE
)

cat("Fitur baru: performance_category dan salary_bracket berhasil ditambahkan.\n")

# ── Data asli ─────────────────────────────────────────────────────────────────
head(df6, 10) %>%
  kable(
    caption     = "Tabel 6.1 — Data Asli (sebelum transformasi)",
    col.names   = c("Company ID", "Employee ID", "Salary", "Department",
                    "Performance Score", "KPI Score",
                    "Performance Category", "Salary Bracket"),
    align       = c("c", "c", "r", "l", "r", "r", "l", "l"),
    digits      = 2,
    format.args = list(big.mark = ",")
  ) %>%
  kable_styling(
    bootstrap_options = c("striped", "hover", "condensed", "bordered"),
    full_width        = TRUE,
    font_size         = 13
  ) %>%
  column_spec(1, bold = TRUE, color = "white", background = "#0277BD") %>%
  column_spec(3, color = "#1B5E20", bold = TRUE) %>%
  column_spec(5, color = "#6A1B9A") %>%
  column_spec(7, italic = TRUE, color = "#E65100") %>%
  column_spec(8, italic = TRUE, color = "#0277BD")

# ── Setelah Min-Max Normalization ─────────────────────────────────────────────
head(df_norm[, numeric_cols], 10) %>%
  kable(
    caption   = "Tabel 6.2 — Setelah Min-Max Normalization (rentang [0, 1])",
    col.names = c("Salary (norm)", "Performance Score (norm)", "KPI Score (norm)"),
    align     = c("r", "r", "r"),
    digits    = 4
  ) %>%
  kable_styling(
    bootstrap_options = c("striped", "hover", "bordered"),
    full_width        = FALSE,
    position          = "center",
    font_size         = 13
  ) %>%
  column_spec(1, color = "#1565C0", bold = TRUE) %>%
  column_spec(2, color = "#1B5E20") %>%
  column_spec(3, color = "#E65100")

# ── Setelah Z-Score Standardization ──────────────────────────────────────────
head(df_z[, numeric_cols], 10) %>%
  kable(
    caption   = "Tabel 6.3 — Setelah Z-Score Standardization (satuan standar deviasi)",
    col.names = c("Salary (z)", "Performance Score (z)", "KPI Score (z)"),
    align     = c("r", "r", "r"),
    digits    = 4
  ) %>%
  kable_styling(
    bootstrap_options = c("striped", "hover", "bordered"),
    full_width        = FALSE,
    position          = "center",
    font_size         = 13
  ) %>%
  column_spec(1, color = "#1565C0", bold = TRUE) %>%
  column_spec(2, color = "#1B5E20") %>%
  column_spec(3, color = "#E65100")

p_sal_before <- ggplot(df6, aes(x = salary)) +
  geom_histogram(bins = 20, fill = "#1565C0", alpha = 0.8, color = "white") +
  labs(title = "Salary — Original", x = "Salary", y = "Count") +
  theme_minimal(base_size = 12)

p_sal_after <- ggplot(df_norm, aes(x = salary)) +
  geom_histogram(bins = 20, fill = "#43A047", alpha = 0.8, color = "white") +
  labs(title = "Salary — Normalized [0, 1]", x = "Normalized Salary", y = "Count") +
  theme_minimal(base_size = 12)

p_perf_before <- ggplot(df6, aes(y = performance_score, x = factor(company_id),
                                  fill = factor(company_id))) +
  geom_boxplot(show.legend = FALSE, alpha = 0.8) +
  scale_fill_brewer(palette = "Pastel1") +
  labs(title = "Performance Score — Original", x = "Company", y = "Score") +
  theme_minimal(base_size = 12)

p_perf_after <- ggplot(df_z, aes(y = performance_score, x = factor(company_id),
                                  fill = factor(company_id))) +
  geom_boxplot(show.legend = FALSE, alpha = 0.8) +
  scale_fill_brewer(palette = "Pastel2") +
  labs(title = "Performance Score — Z-Score", x = "Company", y = "Z-Score") +
  theme_minimal(base_size = 12)

pc_sum <- df6 %>%
  group_by(performance_category) %>%
  summarise(n = n()) %>%
  mutate(pct = round(n / sum(n) * 100, 1))

p_bracket <- ggplot(df6, aes(x = salary_bracket, fill = salary_bracket)) +
  geom_bar(show.legend = FALSE, width = 0.6, color = "white") +
  scale_fill_manual(values = c(
    "Low"       = "#EF9A9A",
    "Medium"    = "#FFF176",
    "High"      = "#A5D6A7",
    "Very High" = "#90CAF9"
  )) +
  labs(title = "Salary Bracket Distribution", x = "Bracket", y = "Count") +
  theme_minimal(base_size = 12)

p_perf_cat <- ggplot(pc_sum, aes(x = "", y = n, fill = performance_category)) +
  geom_bar(stat = "identity", width = 1, color = "white") +
  coord_polar("y") +
  geom_text(aes(label = paste0(pct, "%")),
            position = position_stack(vjust = 0.5),
            size = 3.5, color = "white", fontface = "bold") +
  scale_fill_brewer(palette = "Set2", name = "Category") +
  labs(title = "Performance Category Distribution") +
  theme_void(base_size = 12)

grid.arrange(p_sal_before, p_sal_after,
             p_perf_before, p_perf_after,
             p_bracket, p_perf_cat, ncol = 2)

dash_df <- read.csv("dashboard_data.csv")
cat("Total karyawan  :", nrow(dash_df), "\n")

cat("Total perusahaan:", length(unique(dash_df$company_id)), "\n")

head(dash_df, 10) %>%
  kable(
    caption     = "Tabel 7.1 — Preview dashboard_data.csv (10 baris pertama)",
    col.names   = c("Employee ID", "Company ID", "Salary", "Department",
                    "Performance Score", "KPI Score"),
    align       = c("c", "c", "r", "l", "r", "r"),
    digits      = 2,
    format.args = list(big.mark = ",")
  ) %>%
  kable_styling(
    bootstrap_options = c("striped", "hover", "condensed", "bordered"),
    full_width        = FALSE,
    position          = "center",
    font_size         = 13
  ) %>%
  column_spec(2, bold = TRUE, color = "white", background = "#0277BD") %>%
  column_spec(3, color = "#1B5E20", bold = TRUE) %>%
  column_spec(4, italic = TRUE, color = "#4A148C") %>%
  column_spec(5, color = "#424242") %>%
  column_spec(6, bold = TRUE, color = "#E65100")

# ── Loop kategorisasi KPI tier ────────────────────────────────────────────────
kpi_tier <- c()
for (kpi in dash_df$KPI_score) {
  tier <- if (kpi >= 90) {
    "Tier 1 - Elite"
  } else if (kpi >= 75) {
    "Tier 2 - High"
  } else if (kpi >= 60) {
    "Tier 3 - Average"
  } else {
    "Tier 4 - Low"
  }
  kpi_tier <- c(kpi_tier, tier)
}
dash_df$kpi_tier <- kpi_tier

# ── Summary per company ───────────────────────────────────────────────────────
company_kpi_summary <- dash_df %>%
  group_by(company_id) %>%
  summarise(
    N_Employees    = n(),
    Avg_Salary     = round(mean(salary), 2),
    Avg_KPI        = round(mean(KPI_score), 2),
    Top_Performers = sum(KPI_score >= 90),
    .groups        = "drop"
  )

company_kpi_summary %>%
  kable(
    caption     = "Tabel 7.2 — Ringkasan KPI per Perusahaan",
    col.names   = c("Company", "Jml Karyawan", "Avg Salary (IDR)",
                    "Avg KPI", "Top Performers (KPI >= 90)"),
    align       = c("c", "c", "r", "r", "c"),
    format.args = list(big.mark = ",")
  ) %>%
  kable_styling(
    bootstrap_options = c("striped", "hover", "bordered"),
    full_width        = FALSE,
    position          = "center",
    font_size         = 13
  ) %>%
  column_spec(1, bold = TRUE, color = "white", background = "#0277BD") %>%
  column_spec(3, color = "#1B5E20", bold = TRUE) %>%
  column_spec(4, bold = TRUE,
              color = ifelse(company_kpi_summary$Avg_KPI >= 70, "#1B5E20", "#E65100")) %>%
  column_spec(5, bold = TRUE, color = "white",
              background = ifelse(company_kpi_summary$Top_Performers >= 5,
                                  "#1B5E20", "#E53935"))

# ── Distribusi KPI Tier ───────────────────────────────────────────────────────
tier_order   <- c("Tier 1 - Elite", "Tier 2 - High", "Tier 3 - Average", "Tier 4 - Low")
tier_summary <- dash_df %>%
  group_by(kpi_tier) %>%
  summarise(
    Jumlah     = n(),
    Avg_KPI    = round(mean(KPI_score), 2),
    Avg_Salary = round(mean(salary), 2),
    .groups    = "drop"
  ) %>%
  mutate(
    Persentase = round(Jumlah / sum(Jumlah) * 100, 1),
    kpi_tier   = factor(kpi_tier, levels = tier_order)
  ) %>%
  arrange(kpi_tier)

tier_summary %>%
  kable(
    caption     = "Tabel 7.3 — Distribusi Karyawan per KPI Tier (seluruh perusahaan)",
    col.names   = c("KPI Tier", "Jumlah", "Avg KPI", "Avg Salary (IDR)", "Persentase (%)"),
    align       = c("l", "c", "r", "r", "c"),
    format.args = list(big.mark = ",")
  ) %>%
  kable_styling(
    bootstrap_options = c("striped", "hover", "bordered"),
    full_width        = FALSE,
    position          = "center",
    font_size         = 13
  ) %>%
  column_spec(1, bold = TRUE) %>%
  column_spec(2, color = "#1565C0", bold = TRUE) %>%
  column_spec(3, bold = TRUE,
              color = c("#1B5E20", "#388E3C", "#E65100", "#C62828")) %>%
  column_spec(4, color = "#424242") %>%
  column_spec(5, color = "#6A1B9A") %>%
  row_spec(1, background = "#E8F5E9") %>%
  row_spec(4, background = "#FFEBEE")

p_avg_kpi <- ggplot(company_kpi_summary,
                    aes(x = company_id, y = Avg_KPI, fill = company_id)) +
  geom_bar(stat = "identity", width = 0.6, show.legend = FALSE) +
  geom_text(aes(label = round(Avg_KPI, 1)), vjust = -0.4, fontface = "bold", size = 4) +
  scale_fill_brewer(palette = "Set2") +
  labs(title = "Average KPI Score per Company", x = "Company", y = "Avg KPI") +
  theme_minimal(base_size = 13) +
  ylim(0, 100)

p_top <- ggplot(company_kpi_summary,
                aes(x = company_id, y = Top_Performers, fill = company_id)) +
  geom_bar(stat = "identity", width = 0.6, show.legend = FALSE) +
  geom_text(aes(label = Top_Performers), vjust = -0.4, fontface = "bold", size = 4) +
  scale_fill_brewer(palette = "Set1") +
  labs(title = "Top Performers (KPI >= 90) per Company",
       x = "Company", y = "Count") +
  theme_minimal(base_size = 13) +
  ylim(0, max(company_kpi_summary$Top_Performers) * 1.2)

grid.arrange(p_avg_kpi, p_top, ncol = 2)

dept_summary <- dash_df %>%
  group_by(company_id, department) %>%
  summarise(Avg_Salary = round(mean(salary), 2), .groups = "drop")

ggplot(dept_summary, aes(x = department, y = Avg_Salary, fill = company_id)) +
  geom_bar(stat = "identity", position = "dodge", width = 0.7) +
  scale_fill_brewer(palette = "Dark2", name = "Company") +
  labs(
    title    = "Average Salary by Department and Company",
    subtitle = "Grouped Bar Chart",
    x        = "Department",
    y        = "Avg Salary"
  ) +
  theme_minimal(base_size = 12) +
  theme(
    axis.text.x     = element_text(angle = 30, hjust = 1),
    legend.position = "bottom"
  )

p_scatter <- ggplot(dash_df, aes(x = performance_score, y = KPI_score,
                                  color = company_id)) +
  geom_point(alpha = 0.35, size = 1.5) +
  geom_smooth(aes(group = 1), method = "lm", color = "black",
              se = TRUE, linewidth = 1.2) +
  scale_color_brewer(palette = "Set2", name = "Company") +
  labs(
    title    = "Performance Score vs KPI Score",
    subtitle = "Scatter plot with regression line",
    x        = "Performance Score",
    y        = "KPI Score"
  ) +
  theme_minimal(base_size = 13) +
  theme(legend.position = "bottom")

p_sal_tier <- ggplot(dash_df, aes(x = kpi_tier, y = salary, fill = kpi_tier)) +
  geom_boxplot(show.legend = FALSE, alpha = 0.8, outlier.size = 1) +
  scale_fill_manual(values = c(
    "Tier 1 - Elite"   = "#1B5E20",
    "Tier 2 - High"    = "#388E3C",
    "Tier 3 - Average" = "#FBC02D",
    "Tier 4 - Low"     = "#C62828"
  )) +
  labs(title = "Salary Distribution by KPI Tier",
       x = "KPI Tier", y = "Salary") +
  theme_minimal(base_size = 13) +
  theme(axis.text.x = element_text(angle = 20, hjust = 1))

grid.arrange(p_scatter, p_sal_tier, ncol = 2)