AI System for Hotel and Restaurant HoReCa Business Development

Development of an AI system for the hotel and restaurant business HoReCa

HoReCa is a business with thin margins, high operational complexity, and a critical dependence on user experience. AI optimizes occupancy, manages inventory, personalizes service, and assists with pricing.

Revenue Management

Dynamic Pricing for Hotels:

Real-time rate optimization is the foundation of hotel revenue management:

import numpy as np
import pandas as pd
from sklearn.ensemble import GradientBoostingRegressor

class HotelDemandPredictor:
    """Прогноз спроса на гостиничные номера"""

    def build_features(self, date, hotel_data):
        return {
            # Сезонные факторы
            'days_to_arrival': (date - pd.Timestamp.today()).days,
            'day_of_week': date.dayofweek,
            'is_weekend': int(date.dayofweek >= 4),
            'month': date.month,
            'is_holiday': int(date in hotel_data['holidays']),

            # Конкурентная среда
            'avg_competitor_rate': hotel_data['comp_rates'].get(str(date), 0),
            'min_competitor_rate': hotel_data['comp_min_rates'].get(str(date), 0),

            # Исторические паттерны
            'last_year_occupancy': hotel_data['hist_occupancy'].get(str(date), 0.7),
            'booking_pace_7d': hotel_data['current_bookings'] / hotel_data['capacity'],

            # События в городе
            'event_flag': int(any(e['date'] == str(date) for e in hotel_data['events'])),
            'event_size': sum(e.get('attendees', 0) for e in hotel_data['events']
                             if e['date'] == str(date)),
        }

class DynamicPricingEngine:
    def __init__(self, demand_model, min_rate, max_rate, rack_rate):
        self.demand_model = demand_model
        self.min_rate = min_rate
        self.max_rate = max_rate
        self.rack_rate = rack_rate

    def recommend_rate(self, date, current_occupancy, days_ahead, features):
        # Прогноз спроса при текущем тарифе
        predicted_demand = self.demand_model.predict([features])[0]

        # Уровень заполнения относительно компрессии
        if days_ahead < 7 and current_occupancy > 0.85:
            # Высокий спрос, мало времени → повысить
            multiplier = 1.3 + (current_occupancy - 0.85) * 4
        elif days_ahead > 60 and current_occupancy < 0.4:
            # Далеко и мало броней → снизить для стимуляции
            multiplier = 0.75
        else:
            multiplier = 0.9 + predicted_demand * 0.4  # нормальное динамическое ценообразование

        recommended = np.clip(self.rack_rate * multiplier, self.min_rate, self.max_rate)
        return round(recommended / 100) * 100  # округлить до 100 руб

Restaurant inventory management

Ingredient Consumption Estimate:

• Demand for menu items → decomposition into ingredients by recipes - Features: day of the week, weather (hot dishes in the rain), events in the hotel/city - Forecast MAPE: 8–15% for the daily horizon

Procurement management and waste reduction:

• Safety stock is calculated based on the quantile forecast (P90 demand) - FIFO in warehouse accounting + automatic FEFO for perishables - Food waste report: actual consumption vs. theoretical (by recipe × sold dishes)

Waste reduction: In a typical restaurant, 20–30% of food is wasted. AI reduces this to 8–12%.

Personalizing the guest experience

Guest profile:

From PMS (Property Management System) and booking history: - Room preferences: floor, view, temperature, pillows - Dietary restrictions: vegetarian, halal, allergies - Activities: loves SPA, prefers early breakfast - Communication preferences: WhatsApp, e-mail, do not disturb

Pre-arrival automation: 24 hours before check-in → personal message with an upgrade offer based on the P(accept) model.

Chatbot for guests:

LLM + RAG on the hotel knowledge base (FAQ, menu, attractions): - Requests via messenger (WhatsApp, Telegram) or QR code in the room - “I want to book dinner for two at 8:00 PM” → check availability → confirmation - “Where is the nearest ATM?” → search by location database

Optimizing kitchen and operations

Kitchen load forecast:

LSTM on order history → forecast for 15-minute periods → kitchen staff planning: - Peak 7:00 PM–9:00 PM in the restaurant = 3 cooks on the line - Quiet hour 3:00 PM–5:00 PM = 1 cook

Recipe costing and menu engineering:

• Automatic calculation of the cost of dishes based on current prices of ingredients - Menu Engineering matrix: Stars (popular + profitable), Plowhorses, Puzzles, Dogs - ML experiment: testing new menu items on a subset of guests with AB analysis

Development time: 4–7 months for a HoReCa AI platform with Revenue Management, inventory forecasting, and personalization with integration into PMS (Opera, Hestia, Fidelio).